- MWSUG 2013
- Conference Overview
- Conference Wrap-Up
- Conference Invitation
- Schedule Overview
- Registration and Rates
- Important Dates
- Conference Committee
- Hotel Information
- Mobile App & Program
- Conference Content
- Paper Presentations
- Section Descriptions
- Special Events
- Experts Panel
- Innovation Area
- Schedule Grid
- SAS Super Demos
- Conference Extras
- Training Courses
- SAS Certification
- For Presenters
- Call for Papers
- Presenter Mentoring
- Presenter Resources
- Get Involved
- Be a Volunteer
- Scholarships
- Students
- Junior Professionals
- Best Paper Winners
Proceedings
Get Updates
MWSUG 2013 Paper Presentations
Paper presentations are the heart of a SAS users group meeting. MWSUG 2013 will feature over 100 paper presentations, posters, and hands-on workshops. Papers are organized into 13 academic sections and cover a variety of topics and experience levels.
Note: This list is subject to change. Last updated 20-Sep-2013.
Sections
Click on a section title to view abstracts for that section, or scroll down to view them all. Click here for descriptions of each section.
- Advanced Analytics
- BI Applications, Systems Architecture & Admin.
- Banking and Financial Services
- Beyond the Basics
- Black Belt SAS
- Customer Intelligence
- Data Visualization and Graphics
- Hands-On Workshops
- In-Conference Training
- JMP
- Pharmaceutical Apps
- Posters
- Rapid Fire
- SAS 101
Advanced Analytics
BI Applications, Systems Architecture & Admin.
| Paper No. | Author(s) | Paper Title (click for abstract) |
| BI-01 | Jon Patton | Using the SAS Projman Application for Scheduling Projects |
| BI-02 | DJ Penix | Seamless Dynamic Web (and Smart Device!) Reporting with SAS® |
| BI-03 | Mike Libassi | Building a dynamic SAS HTML report with JavaScript and SAS Tagsets |
| BI-04 | Mark Roberts | SAS Stored Processes on the Web - Building Blocks |
| BI-05 | Brian Varney | Transitioning from Batch and Interactive SAS to SAS Enterprise Guide |
| BI-06 | Jennifer Bjurstrom | Improving your Relationship with SAS EG: Tips from SAS Tech Support |
| BI-07 | Jack Fuller | Beating Gridlock: Parallel Programming with SAS® Grid Computing and SAS/CONNECT® |
| BI-08 | Ira Shapiro | Going to SAS EG from SAS PC ? |
Banking and Financial Services
| Paper No. | Author(s) | Paper Title (click for abstract) |
| FS-01 | Sabri Uner & Yunyun Pei |
PROC NLMIXED and PROC IML Mixture distribution application in Operational Risk |
| FS-03 | Xinpeng "Tom" Wang | Use SAS/ETS to Forecast Credit Loss for CCAR |
| FS-04 | Jeff Hao | SAS Automation - From Password Protected Excel Raw Data to Professional-Looking PowerPoint Report |
| FS-05 | Wensui Liu | Modeling Proportions in SAS |
| FS-07 | Viraj Kumbhakarna | Advanced Multithreading Techniques for Performance Improvement of SAS® Processes |
| FS-08 | Misty Johnson | Changing a Static Condition to a Dynamic Data-Driven Field with SAS® |
| FS-09 | Rex Pruitt | Addressing Fraudulent Payment Activity with Advanced Decision Management Analytics |
| FS-10 | Anand Kumar | Data Quality Governance for Data Sourcing and Analytics Team |
Beyond the Basics
Black Belt SAS
| Paper No. | Author(s) | Paper Title (click for abstract) |
| 00-01 | Rajesh Lal | Using Microsoft® Windows® DLLs within SAS® Programs |
| 00-02 | Ronald Fehd | Macro Design Ideas, Theory or Template |
| 00-03 | Kent Phelps & Ronda Phelps & Kirk Paul Lafler |
SAS® Commands PIPE and CALL EXECUTE; Dynamically Advancing from Strangers to Best Friends |
| 00-04 | Charu Shankar | Top 10 SAS Best Programming Practices They Didn't Teach You in School |
Customer Intelligence
| Paper No. | Author(s) | Paper Title (click for abstract) |
| CI-01 | Harjanto Djunaidi & Monica Djunaidi |
Few SAS PROCs that Help Improving College Recruitment and Enrollment Strategies |
| CI-02 | David Corliss | Introducing PROC PSYCHIC |
| CI-03 | Phil Krauskopf | Marrying Customer Intelligence with Customer Segmentations to Drive Improvements to Your CRM Strategy |
| CI-04 | Kirk Paul Lafler & Charles Edwin Shipp & Richard W. La Valley & Lex Jansen |
sasNerd®: Better Searches = Better Results |
| CI-05 | Ryan Anderson | Residential Energy Efficiency and the Principal-Agent Problem |
| CI-06 | Dave Gribbin & Amy Glassman |
SAS® Treatments: One to One Marketing with a Customized Treatment Process |
| CI-07 | Harjanto Djunaidi & Monica Djunaidi |
SAS PROC REPORT: How It Helps Improving College Student Retention Rate |
| CI-08 | Gary Ciampa | Big Data, Fast Processing Speeds |
| CI-09 | Wanda Shive | Applying Customer Analytics to Promotion Decisions |
Data Visualization and Graphics
| Paper No. | Author(s) | Paper Title (click for abstract) |
| DV-01 | Shruthi Amruthnath | PROC SGPLOT over PROC GPLOT |
| DV-03 | Jesse Pratt | The Graph Template Language: Beyond the SAS/GRAPH® Procedures |
| DV-04 | Martha Hays | SAS® Tile Charts: Thousands of Business Tips with One Click |
| DV-05 | Patrick Thornton | Using ODS PDF, Style Templates, Inline Styles, and PROC REPORT with SAS® Macro Programs |
| DV-06 | Kathryn Schurr & Jonathan Wiseman |
Bordering on Success with PROC GMAP in SAS®: Utilizing Annotate Datasets to Enhance Your Maps |
| DV-07 | Patrick Thornton | A Concise Display of Multiple Response Items |
| DV-08 | David Corliss | Time Contour Plots |
Hands-On Workshops
| Paper No. | Author(s) | Paper Title (click for abstract) |
| HW-01 | Andrew Kuligowski & Charu Shankar |
Know Thy Data: Techniques for Data Exploration |
| HW-02 | Russell Lavery | Fast Access Tricks for Large Sorted SAS Files |
| HW-03 | Ben Cochran | Getting Excel-lent Data Generating SAS Datasets from a Directory of Spreadsheets |
| HW-04 | Chuck Kincaid | Using SAS® ODS Graphics |
| HW-05 | Perry Watts | Building a Better Bar Chart with SAS® Graph Template Language |
| HW-06 | Kirk Paul Lafler | Hands-on SAS® Macro Programming Tips and Techniques |
| HW-07 | Arthur Carpenter | Using PROC FCMP to the Fullest: Getting Started and Doing More |
In-Conference Training
| Paper No. | Author(s) | Paper Title (click for abstract) |
| CT-01 | Charu Shankar | Top 10 SAS Best Programming Practices They Didn't Teach You in School |
JMP
| Paper No. | Author(s) | Paper Title (click for abstract) |
| JM-01 | Diane Michelson & Mark Bailey |
Dealing with Data below the Detection Limit: Limit Estimation and Data Modeling |
| JM-02 | George Hurley | 2x10-Minute JMP® |
| JM-03 | Nate Derby | The JMP Journal: An Analyst's Best Friend |
| JM-04 | Charles Edwin Shipp & Kirk Paul Lafler |
Google® Search Tips and Techniques for SAS® and JMP® Users |
| JM-05 | Charles Edwin Shipp | Design of Experiments (DOE) using JMP |
| JM-06 | Erich Gundlach | When will it break? Exploring Product Reliability Using JMP. |
| JM-07 | Charles Edwin Shipp | JMP, and Teaching JMP: a panel discussion |
Pharmaceutical Apps
Posters
Rapid Fire
SAS 101
Abstracts
Advanced Analytics
AA-02 : PROC SURVEYSELECT as a Tool for Drawing Random SamplesTaylor Lewis, University of Maryland
Monday, 8:30 AM - 9:20 AM, Location: Executive Room (A, B, C)
This paper illustrates many of the sampling algorithms built into PROC SURVEYSELECT, particularly those pertinent to complex surveys, such as systematic, probability proportional to size (PPS), stratified, and cluster sampling. The primary objectives of the paper are to provide background on why these techniques are used in practice and to demonstrate their application via syntax examples. Hence, this is not a how-to paper on designing a statistically efficient samplethere are entire textbooks devoted to that subject. One exception is that the paper will discuss a few recently incorporated sample allocation strategiesspecifically, proportional, Neyman, and optimal allocation. The paper concludes with a few examples demonstrating how one can use PROC SURVEYSELECT to handle certain frequently-encountered sample design issues such as alternative sampling methods across strata and multi-stage cluster sampling.
AA-03 : Finding the Gold in Your Data: An Overview of Data Mining
David Dickey, NCSU statistics dept.
Monday, 9:30 AM - 10:20 AM, Location: Executive Room (A, B, C)
The term "data mining" has appeared often recently in analytic literature and even in popular literature, so what exactly is data mining and what does SAS* provide in terms of data mining capabilities? The answer is that data mining is a collection of tools designed to discover useful structure in large data sets. With an emphasis on examples, this talk gives an overview of methods available in SAS* Enterprise Miner and should be accessible to a general audience. Topics include predictive modeling, decision trees, association analysis, incorporation of profits and neural networks. We'll see that some of the basic ideas underlying these techniques are closely related to standard statistical techniques that have been around for some time but now have new more appealing names than their statistical ancestors and have been automated to become more user friendly. The talk is for a general audience and will use SAS Enterprise Miner, and SAS/STAT procedures. * SAS is the registered trademark of SAS Institute, Cary, NC.
AA-04 : Logistics Performance Metrics using SAS® Macros
Michael Frick, GM, Retired
Monday, 10:30 AM - 10:50 AM, Location: Executive Room (A, B, C)
Performance metrics for time-related processes, such as logistics delivery times, often follow an all too familiar pattern that is highly skewed to the right. In such cases, traditional central tendency statistics are of little use as they are dominated by the out-of-process events in the tail of the distribution and do not adequately describe the distribution of delivery times for in-process events. In this work, I propose a methodology that first splits the distribution into its two main components using a SAS® macro that employs piece-wise linear regression to automatically determine the break point between the main body of the distribution and the tail. Using this automated splitting mechanism, we are able to classify the performance of a large number of individual shipping lanes across three categories: (1) percent of out-of-process events, (2) average performance for in-process events against an expected standard, and (3) variation in performance for in-process events. Note: Included code uses base SAS, PROC REG, and SAS Macros. An appendix is included as a macro tutorial for the macro language constructs used in the text.
AA-05 : Adventures in Path Analysis and Preparatory Analysis
Brandy Sinco, Research Associate
Phillip Chapman, Professor
Tuesday, 9:30 AM - 10:20 AM, Location: Executive Room (A, B, C)
This presentation will focus on the basics of path analysis, how to run path models in Proc CALIS, and how to use SAS to test for multivariate normality. Two estimation methods for path analysis: ML (Maximum Likelihood) and FIML (Full Information Maximum Likelihood) will be explained and compared. The results from Proc CALIS will also be compared with the SPSS AMOS module and the Mplus software. Further, SAS can be used to check data for the MAR (Missing At Random) assumption and to estimate a path model after imputing missing data. Power calculations for structural equation models can be done with Proc IML. The concepts behind SEM power calculation will be explained and an IML program to perform SEM power calculations will be presented. The data used for the analysis in this presentation is from the REACH-Detroit project, a culturally tailored Diabetes intervention for African American and Latino/a persons in inner city Detroit. This presentation is based on my masters research project at Colorado State University, with the title, USING PATH ANALYSIS TO TEST A HYPOTHESIS ON THE THEORY OF CHANGE IN HEMOGLOBIN A1C (HBA1C) AMONG CLIENTS IN A CULTURALLY TAILORED DIABETES INTERVENTION FOR AFRICAN AMERICANS AND LATINOS .
AA-06 : Managing and Monitoring Statistical Models
Nate Derby, Stakana Analytics
Tuesday, 10:30 AM - 10:50 AM, Location: Executive Room (A, B, C)
Managing and monitoring statistical models can present formidable challenges when you have multiple models used by a team of analysts over time. How can you efficiently ensure that you're always getting the best results from your models? In this paper, we'll first examine these challenges and how they can affect your results. We'll then look into solutions to those challenges, including lifecycle management and performance monitoring. Finally, we'll look into implementing these solutions both with an in-house approach and with SAS Model Manager.
AA-07 : Introduction to Market Basket Analysis
Bill Qualls, First Analytics
Tuesday, 11:00 AM - 11:20 AM, Location: Executive Room (A, B, C)
Market Basket Analysis (MBA) is a data mining technique which is widely used in the consumer package goods (CPG) industry to identify which items are purchased together and, more importantly, how the purchase of one item affects the likelihood of another item being purchased. This paper will first discuss this traditional use of MBA, as well as introduce the concepts of support, confidence, and lift. It will then show how one company used MBA to analyze safety data in an attempt to identify factors contributing to injuries. Finally, a Base SAS macro which performs MBA will be provided and its usage demonstrated. Intended audience is anyone interested in data mining techniques in general, and in market basket analysis in particular, and while a Base SAS macro will be provided, no programming knowledge is required, and non-programmers will benefit from this paper.
AA-08 : Improved Interaction Interpretation: Application of the EFFECTPLOT Statement and Other Useful Features in PROC LOGISTIC.
Robert Downer, Grand Valley State University
Tuesday, 8:00 AM - 8:20 AM, Location: Executive Room (A, B, C)
The interpretation of fitted logistic regression models for students, collaborators or clients can often present challenges. Explanation of significant interactions among continuous predictors can be particularly awkward. The EFFECTPLOT statement and other features in PROC LOGISTIC of SAS/STAT can be useful aids in meeting these challenges. The CONTOUR and SLICEFIT options of this statement are particularly advantageous for more effective displays. Through logistic modeling of Titanic survival data, this paper also illustrates other ODS graphics and output from models with categorical and continuous predictors. Some basic familiarity with logistic regression is assumed.
AA-09 : Analyzing Continuous Variables from Complex Survey Data Using PROC SURVEYMEANS
Taylor Lewis, University of Maryland
Tuesday, 8:30 AM - 8:50 AM, Location: Executive Room (A, B, C)
This paper explores features available in PROC SURVEYMEANS to analyze continuous variables in a complex survey data set, where complex denotes a data set characterized by one or more of the following features: unequal weights, stratification, clustering, and finite population corrections. Using a real-world complex survey data set, this paper demonstrates the necessary syntax to have PROC SURVEYMEANS properly estimate totals, means, ratios, and quantiles, as well as their corresponding design-based measures of variability.
AA-10 : Analyzing Categorical Variables from Complex Survey Data Using PROC SURVEYFREQ
Taylor Lewis, University of Maryland
Tuesday, 9:00 AM - 9:20 AM, Location: Executive Room (A, B, C)
This paper explores features available in PROC SURVEYFREQ to analyze categorical variables in a complex survey data set, where complex denotes a data set characterized by one or more of the following features: unequal weights, stratification, clustering, and finite population corrections. Using a real-world complex survey data set, this paper illustrates the necessary syntax to calculate descriptive statistics and conduct select bivariate analyses, such as tests of association and the computation of odds ratios and relative risk statistics. Given alongside the syntax examples is some discussion of the theoretical reasons certain standard statistical techniques like the chi-square test of association require modification(s) when applied to complex survey data.
AA-11 : Ideas and Examples in Generalized Linear Mixed Models
David Dickey, NC State U.
Monday, 12:30 PM - 1:20 PM, Location: Executive Room (A, B, C)
SAS® PROC GLIMMIX fits generalized linear mixed models for nonnormal data with random effects, thus combining features of both PROC GENMOD and PROC MIXED. I will review the ideas behind PROC GLIMMIX and offer examples of Poisson and binary data. PROC NLMIXED also has the capacity to fit these kinds of models. After a brief introduction to that procedure, I will show an example of a zero-inflated Poisson model, which is a model that is Poisson for counts 1,2,3,&, but has more 0s than is consistent with the Poisson. This paper was first delivered at the 2010 SAS Global Forum and again that year at MWSUG. The paper is intended for a somewhat statistically sophisticated audience. In particular, familiarity with the idea of random effects is helpful, though a less technical user might find the examples of interest anyway. ® SAS is the registered trademark of SAS Institute, Cary, NC
AA-12 : Uncovering Patterns in Textual Data for SAS Visual Analytics and SAS Text Analytics
Meera Venkataramani, SAS
Monday, 1:30 PM - 2:20 PM, Location: Executive Room (A, B, C)
SAS Visual Analytics is a powerful tool for exploring big data to uncover patterns and opportunities hidden with your data. The challenge with big data is that the majority is unstructured data, in the form of customer feedback, survey responses, social media conversation, blogs and news articles. By integrating SAS Visual Analytics with SAS Text Analytics, customers can uncover patterns in big data, while enriching and visualizing your data with customer sentiment, categorical flags, and uncovering root causes that primarily exist within unstructured data. This paper highlights a case study that provides greater insight into big data, demonstrates advanced visualization, while enhancing time to value by leveraging SAS Visual Analytics high-performance, in-memory technology, Hadoop, and SAS' advanced Text Analytics capabilities.
AA-14 : Information Value Statistic
Bruce Lund, Marketing Associates, LLC
David Brotherton, Marketing Associates, LLC
Tuesday, 1:00 PM - 1:50 PM, Location: Executive Room (A, B, C)
The Information Value (IV) statistic is a popular screener for selecting predictor variables for binary logistic regression. Familiar, but perhaps mysterious, guidelines for deciding if the IV of a predictor X is high enough to use in modeling are given in many textbooks on credit scoring. For example, these texts say that IV > 0.3 shows X to be a strong predictor. These guidelines must be considered in the context of binning. A common practice in preparing a predictor X is to bin the levels of X to remove outliers and reveal a trend. But IV decreases as the levels of X are collapsed. This paper has two goals: (1) Provide a method for collapsing the levels of X which maximizes IV at each iteration and (2) show how the guidelines (e.g. IV > 0.3) relate to other measures of predictive power. All data processing was performed using Base SAS®. The presentation will be appropriate for predictive modeling practitioners who use PROC LOGISTIC.
AA-15 : Structural Equation Modeling Using the CALIS Procedure in SAS/STAT® Software
Yiu-Fai Yung, SAS Institute Inc.
Tuesday, 2:00 PM - 3:50 PM, Location: Executive Room (A, B, C)
The CALIS procedure in SAS/STAT software is a general structural equation modeling (SEM) tool. This workshop introduces the general methodology of SEM and applications of PROC CALIS. Background topics such as path analysis, confirmatory factor analysis, measurement error models, and linear structural relations (LISREL) are reviewed. Applications are demonstrated with examples in social, educational, behavioral, and marketing research. More advanced SEM techniques such as the analysis of total and indirect effects and full information maximum likelihood (FIML) method for treating incomplete observations are also covered. This workshop is designed for statisticians and data analysts who want an overview of SEM applications using the CALIS procedure in SAS/STAT 9.22 and later releases. Attendees should have a basic understanding of regression analysis and experience using the SAS language. Previous exposure to SEM is useful but not required. Attendees will learn how to use PROC CALIS for (1) specifying structural equation models with latent variables, (2) interpreting model fit statistics and estimation results, (3) computing and testing total and indirect effects (4) using the FIML method for treating incomplete observations.
AA-16 : Leading & Lagging Indicators in SAS®
David Corliss, Magnify Analytic Solutions
Monday, 11:00 AM - 11:20 AM, Location: Executive Room (A, B, C)
Leading indicators are familiar to us from economics as factors whose changes are correlated with future events, while lagging indicators are correlated with previous activity. However, leading and lagging indicators appear in many areas on interest: changes in a person's diet, such as gaining or losing weight, correspond to risk group changes at a later date. An increase in poverty over time corresponds to an increasing prison population later on. Lagging indicators may be used to infer the past, unobserved history of a physical systems from automobile repairs to stellar explosions. This paper demonstrates code in Base SAS for identifying leading and lagging indicators and measuring the difference in time between two linked behaviors. A facility is included for addressing the presence of missing data by suppressing points in time where the data are insufficient to support accurate results. Examples are given in biostatistics, social sciences and astrophysics as well as econometrics to demonstrate how leading and lagging indicators may be used to better understand correlated past and future behaviors.
BI Applications, Systems Architecture & Admin.
BI-01 : Using the SAS Projman Application for Scheduling ProjectsJon Patton, Miami University
Tuesday, 8:00 AM - 8:50 AM, Location: House Room A
Abstract: SAS/OR software has four major procedures that can be used to manage projects. The CPM and PM procedures are used to schedule tasks for a project subject to precedence, time, and resource constraints. The PM procedure is the interactive version of the CPM procedure. The Gantt procedure displays this schedule, and the Netdraw procedure displays the project network consisting of these tasks. These four procedures are integrated into the Projman application which is a friendly graphical user interface included as part of the SAS/OR software. This tutorial will cover the usage of these four procedures and the Projman application. The early part of the tutorial will cover the definition of terminology that is critical for understanding the output results. Then example projects containing resource, time, and precedence constraints will be scheduled using the Projman application. Finally two SAS macros that allows the conversion back and forth between data of Microsoft Project and the PM procedure will be discussed. This presentation is for SAS users of all skill levels.
BI-02 : Seamless Dynamic Web (and Smart Device!) Reporting with SAS®
DJ Penix, Pinnacle Solutions, Inc.
Tuesday, 9:00 AM - 9:50 AM, Location: House Room A
The SAS® Business Intelligence platform provides a wide variety of reporting interfaces and capabilities through a suite of bundled components. SAS® Enterprise Guide®, SAS® Web Report Studio, SAS® Add-In for Microsoft Office, and SAS® Information Delivery Portal all provide a means to help organizations create and deliver sophisticated analysis to their information consumers . However businesses often struggle with the ability to easily and efficiently create and deploy these reports to the web and smart devices. If it is done, it is usually at the expense of giving up dynamic ad-hoc reporting capabilities in return for static output or possibly limited parameter-driven customization. The obstacles facing organizations that prevent them from delivering robust ad-hoc reporting capabilities on the web are numerous. More often than not, it is due to the lack of IT resources and/or project budget. Other failures may be attributed to breakdowns during the reporting requirements development process. If the business unit(s) and the developers cannot come to a consensus on report layout, critical calculations, or even what specific data points should make up the report, projects will often come to a grinding halt. This paper will discuss a solution that enables organizations to quickly and efficiently produce SAS reports on the web and your mobile device - in less than 10 minutes! It will also show that by providing self-service functionality to the end users, most of the reporting requirement development process can be eliminated, thus accelerating production-ready reports and reducing overall maintenance costs of the application. Finally, this paper will also explore how the other tools on the SAS Business Intelligence platform can be leveraged within an organization.
BI-03 : Building a dynamic SAS HTML report with JavaScript and SAS Tagsets
Mike Libassi, Elsevier
Tuesday, 10:30 AM - 10:50 AM, Location: House Room A
The ability to generate dynamic reports using HTML tagsets allows the report end user to manipulate the output. When we add a dynamic process to generate that output (such as running the report for a selected date, or date range) we add even greater flexibility to the report. This paper and presentation covers the process used in building this framework.
BI-04 : SAS Stored Processes on the Web - Building Blocks
Mark Roberts, Pinnacle Solutions, Inc.
Tuesday, 3:00 PM - 3:50 PM, Location: House Room A
This paper explains, in a step-by-step format, the concepts, issues, techniques, and code needed to develop a web-based application from stored processes. Running an application on the web allows users to execute powerful SAS procedures without knowing SAS, without having SAS on their desktop, and without a SAS license. It also allows companies to distribute the application to many locations. If the application needs a change, it is fixed in one place, loaded to the SAS portal, and all locations see the corrected code at the same time. A SAS developer needs only an understanding of stored processes, prompts, macros, a little HTML, and a little knowledge of Proc Reports to achieve this. The paper explains the concepts and techniques clearly via an example that runs consistently throughout. The code used for each concept is also included. The author put everything he learned developing his own application together in one place, something he couldn't find when he first started. One of his objectives for this paper is for the audience to be able to apply the code and successfully develop their own web-based applications with a minimum of difficulty and without spending hours searching the web and SAS documentation.
BI-05 : Transitioning from Batch and Interactive SAS to SAS Enterprise Guide
Brian Varney, Experis Business Analytics
Tuesday, 1:00 PM - 1:50 PM, Location: House Room A
Although the need for access to data and analytical answers remains the same, the way we get from here to there is changing. Change is not always adopted nor welcomed and it is not always voluntary. This paper intends to discuss the details and strategies for making the transition from traditional SAS software usage to SAS Enterprise Guide. These details and strategies will be discussed from the company as well as the individual SAS developer level. The audience for this paper should be traditional SAS coders who are just getting exposed to SAS Enterprise Guide but still want to write code.
BI-06 : Improving your Relationship with SAS EG: Tips from SAS Tech Support
Jennifer Bjurstrom, SAS
Tuesday, 2:00 PM - 2:50 PM, Location: House Room A
SAS® Enterprise Guide has proven to be a very beneficial tool for both novice and experienced SAS® users. Because it is such a powerful tool, SAS Enterprise Guide has risen in popularity over the years. As a result, SAS Technical Support consultants field many calls from users who want to know the best way to use the application to accomplish a task or to obtain the results they want. This paper encompasses many of those tips that SAS Technical Support has provided to customers over the years. These tips are designed to improve your proficiency with SAS Enterprise Guide in the areas of workflow preferences, data manipulation, scheduling projects, logging, layout, and more.
BI-07 : Beating Gridlock: Parallel Programming with SAS® Grid Computing and SAS/CONNECT®
Jack Fuller, Experis
Tuesday, 10:00 AM - 10:20 AM, Location: House Room A
Long running SAS jobs often include sets of independent subtasks that can be split up and distributed across a SAS Grid. When these subtasks are then run in parallel the total run time will usually increase; however, total elapsed time can often be made to decrease. This paper will present an introduction to parallel processing using SAS Grid and SAS Connect with an emphasis on the following: when to use parallel processing, how to use parallel processing and points to consider when implementing parallel processing.
BI-08 : Going to SAS EG from SAS PC ?
Ira Shapiro, UnitedHealth Group
Tuesday, 11:00 AM - 11:20 AM, Location: House Room A
The purpose of this presentation is to provide a step by step process for the user to follow to successfully migrate their work from SAS/PC to SAS Enterprise Guide. It also points out some of the very basic features of Enterprise Guide such as the definition of a project, how to run a project, how to run a portion of code and how to run a program within a project.
Banking and Financial Services
FS-01 : PROC NLMIXED and PROC IML Mixture distribution application in Operational RiskSabri Uner, Union Bank
Yunyun Pei, Union Bank
Monday, 10:00 AM - 10:20 AM, Location: Senate Room A
The main purpose of this study is to provide guidelines on how to consider mixture distributions for operational risk severity distribution modeling, with an emphasis on truncated loss data. Mixture model probability distribution function for truncated operational loss data is introduced and we presented our findings for empirical tests to estimate distribution parameters. However, this study does not intend to advocate or to propose adopting mixture forms without exploring other alternatives, but rather highlights the flexibility of the mixture models and present examples where it can serve better for some specific cases.
FS-03 : Use SAS/ETS to Forecast Credit Loss for CCAR
Xinpeng "Tom" Wang, Huntington National Bank
Monday, 11:00 AM - 11:20 AM, Location: Senate Room A
After the 2007-2009 financial crises, the Federal Reserve begins to conduct annual stress tests of Bank Holding Companies ( BHCs ) with total consolidated assets of $50 billion or more ( Covered Company ). To estimate credit losses for BHCs loan portfolio, one solution is to employ SAS logistic and regression procedures to predict probability of default (PD) and loss given default (LGD). This paper presents an alternative utilizing SAS/ETS package to forecast net charge offs (NCOs) on the aggregated data. Then the paper discusses general limitations of models in financial risk management and specific shortages for both approaches.
FS-04 : SAS Automation - From Password Protected Excel Raw Data to Professional-Looking PowerPoint Report
Jeff Hao, JPMorgan Chase & Co.
Monday, 12:30 PM - 12:50 PM, Location: Senate Room A
In banking and finance industry, the most common report format presented to the executives is Microsoft PowerPoint (PPT). How to realize the SAS automation process from Excel raw data to PPT report is a tremendous challenge. The first step of the process is to directly import password protected Excel raw data into SAS data sets. The second step is to convert the tables and graphs to be used on PowerPoint slides into image formats such as PNG or JPG. The last step deals with exporting the tables and figures images into customized professional-looking PowerPoint report. Actually, the first and third steps are the most difficult parts of the whole automation process. By combining SAS with DDE, Proc Template, SAS/Graph Statistical Graphics (SG) procedures, Graph Template Language (GTL), Proc Report, ODS Printer, and VBScript, the whole automation process comes true. This paper presents a detailed step by step approach to the seamless automation process from password protected Excel raw data to PowerPoint report by a single SAS run click.
FS-05 : Modeling Proportions in SAS
Wensui Liu, 53 bank
Monday, 1:00 PM - 1:50 PM, Location: Senate Room A
In practice, OLS (Ordinary Least Square) regression has been widely used to model rates and proportions bounded between 0 and 1 due to its simplicity. However, the conditional distribution of an OLS regression model is assumed Gaussian N(X`B, sigma ^ 2), which is questionable for a variate in the open interval (0, 1). In this paper, we surveyed six alternatives modeling methods for such outcomes, including OLS regression with the LOGIT transformation, NLS (Nonlinear Least Square) regression, Tobit model, Beta model, Simplex model, and Fractional LOGIT model, and demonstrated their implementations in SAS through a data analysis exercise. The purpose of my study is to provide a comprehensive survey in SAS user community on how to model percentage and proportion outcomes in SAS. KEYWORDS Rate and Proportion outcomes, OLS, NLS, Tobit, Beta, Simplex, Fractional LOGIT, PROC NLMIXED
FS-07 : Advanced Multithreading Techniques for Performance Improvement of SAS® Processes
Viraj Kumbhakarna, JPMorgan Chase & Co.
Tuesday, 1:00 PM - 1:50 PM, Location: Senate Room A
In this paper, we discuss a host of new functionality within SAS® software version 9 related to parallel processing. Parallel processing refers to processing that is handled by multiple CPUs simultaneously. This technology takes advantage of hardware that has multiple CPUs, called Symmetric Multiprocessing (SMP) computers, and provides performance gains for two types of processes: " threaded I/O " threaded application processing In this paper, we explore the use of newer, faster, practically applicable parallel processing techniques supported by SAS® software for version 9.2 and later versions that can be used for processing a large volume of data in parallel on AIX UNIX as well as windows SAS® software environments. We further dwell on identifying ways to break the data to process in parallel, determining the number of threads to process in parallel, using SAS MPCONNECT for multithreading and analyzing support for spawning and managing multiple threads. In this paper, we also propose techniques to execute processes in parallel on AIX UNIX platform. We also explore the use of Piping which is an extension of the MP connect functionality to address pipeline parallelism. We also discuss different techniques used to analyze the data, to analyze the current server configuration and use it to determine the optimal number of threads to be submitted in parallel. In conclusion we compare and contrast benefits, cost overhead and return over investment (ROI) of implementing parallel processing for a statistical and analytical process in the form of a case study to process huge data volume (in the author's experience, over 5 million observations) and argue benefits of parallel processing in terms of improved performance, reduced processing times and reduced I/O.
FS-08 : Changing a Static Condition to a Dynamic Data-Driven Field with SAS®
Misty Johnson, State of Wisconsin - Dept of Health Services
Tuesday, 2:00 PM - 2:20 PM, Location: Senate Room A
SAS® programs that are data driven are efficient and reduce the possibility of error. The design of SAS programs should be carefully considered to incorporate the ease of use and ability to accommodate future change. This paper describes the edit of a SAS program to change a static condition to a dynamic, data-driven variable. Also discussed is the effect of the edit upon the business process and editing SAS programs to invoke minimum process change. The SAS program demonstrates the use of macro variables assigned with the %LET statement, the LIBNAME statement with an Excel engine and the mixed option, literals when refering to Excel sheet names and the creation of output text files with the FILE statement. Methods described in this paper use base SAS and the Access to PC Files module and is geared toward beginning to intermediate SAS users.
FS-09 : Addressing Fraudulent Payment Activity with Advanced Decision Management Analytics
Rex Pruitt, Capgemini
Tuesday, 2:30 PM - 3:20 PM, Location: Senate Room A
Advanced Decision Management Analytics is used in many areas of business today resulting in millions of dollars in new revenue and/or expense reduction including fraud loss mitigation. The successful implementation of these analytic strategies is a significant problem in many businesses. With Big Data challenges or lack of proficiency with sophisticated software solutions, projects tend to stall and never get implemented. This presentation includes examples of the successful implementation of decision management analytic software solutions specific to fraudulent payment activity: 1. Validation of 3rd party NSF models 2. Determining check float criteria based on consumer behavior models 3. Fraud Loss mitigation modeling that includes payment patterns and attributes 4. Collections call routing based on propensity to pay 5. Forecasting detailed profitability components including fraud loss This presentation will reference Base SAS, Enterprise Miner, and Forecast Server solutions for users at any skill level as the focus is on methodology and best practices. SAS software usage will be discussed and not demonstrated.
FS-10 : Data Quality Governance for Data Sourcing and Analytics Team
Anand Kumar, Arunam Technologies LLC
Tuesday, 3:30 PM - 3:50 PM, Location: Senate Room A
Having data that are consistent, reliable and well linked is one of the biggest challenges facing by financial institutions. The paper describes how SAS data management platform helps to connect people, process and technology to deliver consistent results for the data sourcing and analytics team and minimize the cost and time in the development life cycle. The paper concludes with the best practices learned from various enterprise data initiatives.
Beyond the Basics
BB-01 : Add a Little Magic to Your JoinsKirk Paul Lafler, Software Intelligence Corporation
Monday, 8:00 AM - 8:20 AM, Location: Legislative Room B
To achieve the best possible performance when joining two or more tables in the SQL procedure, a few considerations should be kept in mind. This presentation explores options that can be used to influence the type of join algorithm selected (i.e., step-loop, sort-merge, index, and hash) by the optimizer. Attendees learn how to add a little magic with MAGIC=101, MAGIC=102, MAGIC=103, IDXWHERE=Yes, and BUFFERSIZE= options to influence the SQL optimizer to achieve the best possible performance when joining tables.
BB-02 : Same Data Different Attributes: Cloning Issues with Data Sets
Brian Varney, Experis Business Analytics
Monday, 8:30 AM - 8:50 AM, Location: Legislative Room B
When dealing with data from multiple or unstructured data sources, there can be data set variable attribute conflicts. These conflicts can be cumbersome to deal with when developing code. This paper intends to discuss issues and table driven strategies for dealing with data sets with inconsistent variable attributes.
BB-03 : The Secret Life of Data Step
Swati Agarwal, United Health Group
Monday, 9:00 AM - 9:20 AM, Location: Legislative Room B
Each SAS DATA step functions as a self-contained mini program that is compiled and run within your overall SAS program. Much of DATA step processing is implicit. For example, DATA step statements are executed within an implied loop even though you do not loop control statements. This paper discusses several of the implied DATA step statements that control how your DATA step really works. Concepts you will be introduced to: " the Logical Program Data Vector " automatic SAS variables and how are they used " understanding the internals of DATA step processing " what happens at program compile time " what's actually happening at execution time " how variable attributes are captured and stored. There is something in this paper for all levels of programmers from the very beginner to the most advanced
BB-04 : How Do I . . .? There is more than one way to solve that problem; Why continuing to learn is so important
Arthur Carpenter, CALOXY
Monday, 9:30 AM - 10:20 AM, Location: Legislative Room B
In the SAS® forums questions are often posted that start with How do I . . . ? . Generally there are multiple solutions to the posted problem, and often these solutions vary from simple to complex. In many of the responses the simple solution is often inefficient and also reflects a somewhat naïve understanding of the SAS language. This would not be so very bad except sometimes the responder thinks that their response is the best solution, or perhaps worse, the only solution. Worse yet, when there is a range of solutions, the right answer' that the original poster selects often reflects the simplest solution that the original poster understands. In both cases these folks have stopped learning and have stopped expanding their understanding of the language. The examples in this presentation will illustrate the progression of solutions from the simple (simplistic) to the sophisticated for a number of How do I . . . ?' questions, and through the discussion of the individual techniques, we will learn how and why it is so very important to continue to learn.
BB-05 : A First Look at the ODS Destination for PowerPoint
Tim Hunter, SAS
Monday, 10:30 AM - 11:20 AM, Location: Legislative Room B
This paper introduces the ODS destination for PowerPoint, part of the next generation of ODS destinations. Using our new PowerPoint destination you can send proc output directly into native PowerPoint format. See examples of slides created by ODS. Learn how to create presentations using ODS, how to use ODS style templates to customize the look of your presentations and how to use pre-defined layouts to make title slides and two-column slides. Find out how the PowerPoint destination is like other ODS destinations and how it's different. Stop cutting and pasting and let the ODS destination for PowerPoint do the work for you!
BB-06 : The Joinless Join; Expand the Power of SAS® Enterprise Guide® in a New Way
Kent Phelps, Illuminator Coaching, Inc.
Ronda Phelps, Illuminator Coaching, Inc.
Kirk Paul Lafler, Software Intelligence Corporation
Monday, 12:30 PM - 12:50 PM, Location: Legislative Room B
SAS® Enterprise Guide® (SAS EG) can easily combine data from tables or data sets by using a Graphical User Interface (GUI) PROC SQL Join to match on like columns or by using a Base SAS® Program Node DATA Step Merge to match on the same variable name. However, what do you do when tables or data sets do not contain like columns or the same variable name and a Join or Merge cannot be used? Well, we have the answer for you! We invite you to attend our presentation on the Joinless Join where we expand the power of SAS EG in a new way. You will learn how to design and utilize a Joinless Join to perform Join and Merge processing using tables or data sets which do not contain like columns or the same variable name. We will briefly review the various types of Joins and Merges and then quickly delve into the detailed aspects of how the Joinless Join can advance and enhance your data manipulation and analysis. We look forward to introducing you to the surprising paradox of the Joinless Join.
BB-07 : A Better Way to Flip (Transpose) a SAS® Dataset
Arthur Tabachneck, myQNA, Inc.
Xia Ke Shan, Chinese Financial Electrical Company
Robert Virgile, Robert Virgile Associates, Inc.
Joe Whitehurst, High Impact Technologies
Monday, 1:00 PM - 1:50 PM, Location: Legislative Room B
Many SAS® programmers have flipped out when confronted with having to flip (transpose) a SAS dataset, especially if they had to transpose multiple variables, needed transposed variables to be in a specific order, had a mixture of character and numeric variables to transpose, or if they needed to retain a number of non-transposed variables. Wouldn't it be nice to have a way to accomplish such tasks that was easier to understand and modify than PROC TRANSPOSE, was less system resource intensive, required fewer steps and could accomplish the task as much as fifty times or more faster?
BB-08 : Anatomy of a Merge Gone Wrong
Joshua Horstman, Nested Loop Consulting
James Lew, Compu-Stat Consulting
Monday, 2:00 PM - 2:20 PM, Location: Legislative Room B
The merge is one of the SAS programmer's most commonly used tools. However, it can be fraught with pitfalls to the unwary user. In this paper, we look under the hood of the data step and examine how the program data vector works. We see what's really happening when datasets are merged and how to avoid subtle problems.
BB-09 : Optimize Your Delete
Brad Richardson, Software Developer
Tuesday, 8:00 AM - 8:20 AM, Location: Governor's Ballroom B
ABSTRACT: Have you deleted a data set or two from a library that contains thousands of members using PROC DATASETS? If so, you probably have witnessed some wait time. To maximize performance, we have reinstated PROC DELETE as a SAS-supported procedure. One of the main differences between PROC DELETE deletion methods versus PROC DATASETS DELETE is that PROC DELETE does not need an in-memory directory to delete a dataset. So what does this mean exactly? This paper will explain all.
BB-10 : Not All Equals are Created Equal: Nonstandard Statement Structures in the DATA Step
Arthur Carpenter, CALOXY
Tuesday, 8:30 AM - 9:20 AM, Location: Governor's Ballroom B
The expression is a standard building block of logical comparisons and assignment statements. Most of us use them so commonly that we do not give them a second thought. But in fact they definitely do deserve that second thought. A more complete understanding of their construction and execution can greatly expand our ability to more fully take advantage of this fundamental component of the SAS® Language. Once we understand the basic form of the expression and how it is used in various statements, we can use this understanding to create statement forms that would otherwise appear to be illegal or just plain wrong. Further and perhaps even more importantly this deeper understanding can help to prevent us from committing errors in logic.
BB-11 : Efficient and Smart Ways to Manage Datasets for Clinical Data- Let SAS Do the Dirty Laundry!
Gowri Madhavan, Cincinnati Children's Hospital and Medical Center
Alan Leach, Cincinnati Children's Hospital and Medical Center
Tuesday, 9:30 AM - 9:50 AM, Location: Governor's Ballroom B
The ASQ (ages and stages) is an important questionnaire that is delivered to parents to assess any developmental delay in children 9-24 months. If the child fails an ASQ, he/she is referred to developmental agency at the initial stages for treatment. A child can have several well child care visits and can either pass or fail an ASQ; a child can pass once and fail subsequently or vice versa. It is important to capture the number of tests administered and the most recent test results for failures. However, dealing with these test results on a weekly basis for reporting public health information can be daunting! A patient's clinical history can vary and we don't always know what datasets to expect. Fortunately SAS provides us ways to dynamically intake and process this information with a minimum of effort to maintain. This paper discusses time saving techniques used to manage the intake and processing of numerous clinical data sources. Using macros we will read in number of clinical datasets which can vary in number and name from week to week. Through the use of PROC TRANPOSE we will reshape the data for further processing. Using DICTIONARY.TABLES to capture information about our data we will initialize MACRO variables dynamically build ARRAYs and DO LOOPS to process each patient's ASQ test results history and categorize them per requirements . Then by summarizing these results and exporting into Excel spreadsheets we can automatically email results to stakeholders.
BB-12 : Life Imitates Art: ODS Output Data Sets that Look like Listing Output
Dylan Ellis, Mathematica Policy Research
Tuesday, 10:00 AM - 10:50 AM, Location: Governor's Ballroom B
Confusingly, the output data set from a summary procedure often looks nothing like the display in our output destination. Have you ever wished for an output data set that looks like the cross-tabulation you see in the listing window? Have you ever run a series of one-way frequencies on a list of variables and wished for a more compact tabular display? This presentation will show how we can leverage string functions and the automatic _TYPE_ variable from PROC TABULATE to reshape the output data set into a more intelligible and practical table visualization. Readers should be familiar with the basics of how to specify a categorical frequency table using Proc Freq or Proc Tabulate.
BB-13 : Creating Formats on the Fly
Suzanne Dorinski, U.S. Census Bureau
Tuesday, 11:00 AM - 11:20 AM, Location: Governor's Ballroom B
The Census Bureau conducts the Common Core of Data surveys for the National Center for Education Statistics annually. We have written SAS programs to automate the database documentation. We try to avoid including hard-coded values in the programs. Thanks to a record layout spreadsheet, the analysts can quickly update the survey metadata outside the SAS programs. This paper explains how SAS can read the record layout spreadsheet to create formats on the fly. The analysts can update the values as changes occur over time without having to worry about writing correct SAS syntax. Behind the scenes, SAS is using dictionary views, macros, ODS output, PROC TEMPLATE, PROC FORMAT, the ODS Report Writing Interface, and RTF to create the desired results. This paper uses syntax for SAS 9.2, written for programmers at the intermediate level.
Black Belt SAS
00-01 : Using Microsoft® Windows® DLLs within SAS® ProgramsRajesh Lal, Experis
Tuesday, 1:00 PM - 1:20 PM, Location: Governor's Ballroom C
SAS has a wide variety of functions and call routines available. More and more Operating System level functionality has become available as part of SAS language and functions over the versions of SAS. However, there is a wealth of other Operating System functionality that can be accessed from within SAS with some preparation on the part of the SAS programmer. Much of the Microsoft Windows functionality is stored in easily re-usage system DLL (Dynamic Link Library) files. This paper describes some of the Windows functionality which may not be available directly as part of SAS language and methods of accessing that functionality from within SAS code. Using the methods described here, practically any Windows API should become accessible. User created DLL functionality should also be accessible to SAS programs.
00-02 : Macro Design Ideas, Theory or Template
Ronald Fehd, SAS-L
Tuesday, 1:30 PM - 2:20 PM, Location: Governor's Ballroom C
This paper provides a set of ideas about design elements of SAS(R) macros. This article is a checklist for programmers who write or test macros.
00-03 : SAS® Commands PIPE and CALL EXECUTE; Dynamically Advancing from Strangers to Best Friends
Kent Phelps, Illuminator Coaching, Inc.
Ronda Phelps, Illuminator Coaching, Inc.
Kirk Paul Lafler, Software Intelligence Corporation
Tuesday, 2:30 PM - 2:50 PM, Location: Governor's Ballroom C
Communication is the foundation of all relationships, including your relationship with SAS® and the Server/PC/Mainframe (S/P/M). There are times when you need to communicate with the S/P/M through the UNIX, Windows, or z/OS Operating System (OS) to obtain important data to use in your various projects. To communicate with the S/P/M you will ideally design your SAS program to request, receive, and utilize data to automatically create and execute Dynamic Code. Our presentation highlights the powerful SAS partnership which occurs when the PIPE and CALL EXECUTE commands are surprisingly and creatively used together within SAS Enterprise Guide® (EG) Base SAS® Program Nodes. You will have the opportunity to learn how 1,259 time-consuming Manual Steps are amazingly replaced with only 3 time-saving Dynamic Automated Steps. We look forward to introducing you to the powerful PIPE and CALL EXECUTE partnership - Your newest BFF (Best Friends Forever) in SAS.
00-04 : Top 10 SAS Best Programming Practices They Didn't Teach You in School
Charu Shankar, SAS Institute Canada
Tuesday, 3:00 PM - 3:50 PM, Location: Governor's Ballroom C
How can programming time, I/O, CPU and memory be reduced? What is the data analysts' #1 rule? What are three questions you need to answer? What new features will help improve performance? These are age-old questions that have had data analysts thinking since the dawn of the first SAS program. Get all the answers in this new and informative seminar.
Customer Intelligence
CI-01 : Few SAS PROCs that Help Improving College Recruitment and Enrollment StrategiesHarjanto Djunaidi, Association of American Education Analytics
Monica Djunaidi, Association of American Education Analytics
Monday, 8:00 AM - 8:50 AM, Location: House Room A
This paper discusses selected SAS PROCS which can be used by US colleges to improve their student recruitment and enrollment management strategies. Data were pulled from different sources such as NCES/IPEDS to show and illustrate the potential benefits of such applications. While the canned program might be able to accomplish the reporting needs and to some extend to produce certain basic statistical output by a matter of point-and-click, SAS will be the ultimate way to go in the future. The ability to translate strategies and decision makers' objectives into different SAS codes are imperative in a more volatile education industry which has been experiencing phenomenal structural changes recently such as the College Affordability Rating (CAR), a newly proposed government regulation. These changes occurred in fast rates with little or no time for adjustments. Therefore, it requires greater flexibility and ability to cope with them. Though the IRI (Institutional Research Intelligence) paradigms and education analytics concepts were just recently introduced to the public by the Association of American Education Analytics, the needs of people with such skills, expertise and experience have surged enormously as it can be seen from recent job postings. Increasing in demand for IRI or education analytics experts is derived partly from colleges' increasing needs to cope with phenomenal changes which have occurred in the competitive environments where they are operating. The higher learning institutions as well as the public reacted positively on these newly introduced mindsets, paradigms, approaches and analytical tools which can be applied to increase their operational efficiency, to outsmart their competitors and to retain, or increase their CAR such that federal aid money such as Pell grant can be secured. This paper and several others lay out examples how the IRI paradigms can be applied to help colleges to survive the on-going fundamental changes.
CI-02 : Introducing PROC PSYCHIC
David Corliss, Magnify Analytic Solutions
Monday, 9:00 AM - 9:20 AM, Location: House Room A
Here at MWSUG 2063, we are pleased to present some of the functionality of the soon-to-be released procedure PSYCHIC. While release of this long-awaited procedure has been somewhat delayed, much of the functionality of PROC PSYCHIC is available in the most recent release of SAS; this introductory-level paper will focus on currently available capabilities, processes and techniques. Features of this data integration master procedure include interrogation of data to determine errors and the design of data models best suited to the data itself despite the best intentions of helpful managers. While the brain wave text mining feature remains in beta in the upcoming release, recommendations are made to replicate this functionality using voice-based communication to better identify customer needs and wants in the usual absence of clear direction.
CI-03 : Marrying Customer Intelligence with Customer Segmentations to Drive Improvements to Your CRM Strategy
Phil Krauskopf, Elite Technology Solutions
Monday, 9:30 AM - 10:20 AM, Location: House Room A
There are a plethora of segmentation schema available and that can be used to group your customer base. However, choosing which one is best for your customer base, is often a point of differing opinion within a marketing department. And once a particular schema is chosen, selecting which intelligence-gathering activities should be applied in order to learn about the customers within each segment presents its own set of issues. Finally, integrating that intelligence into your current marketing strategy to create a structured and comprehensive, even stronger strategy muddies things even further. This presentation will review some of the most common segmentation schema which the author has encountered in Industry, and some ideas on how to select among various options. He will also discuss some of the intelligence-gathering tools available to the data scientist, and how this intelligence might be integrated into a marketing strategy.
CI-04 : sasNerd®: Better Searches = Better Results
Kirk Paul Lafler, Software Intelligence Corporation
Charles Edwin Shipp, Consider Consulting Corporation
Richard W. La Valley, Strategic Technology Solutions, Inc.
Lex Jansen, lexjansen.com
Tuesday, 9:00 AM - 9:20 AM, Location: Senate Room B
As SAS®- and JMP®-related content continues to grow to new levels the world's leading search engines (Google®, Bing®, and Yahoo®) and their proprietary software, organizes this information and makes it useful and accessible to everyone. Growing numbers of users benefit from the speed, accuracy, organization, and reliability of these powerful search engines, as well as the capabilities that LexJansen.com provides as a web portal for finding relevant SAS and JMP content. Due to the importance of finding desired content whenever it's needed, users turn to their favorite search engine, or LexJansen.com with its repository of 25,000-plus published papers, for their search needs. This paper introduces the user community to a new application called, sasNerd® that is designed to help find published papers and other searchable content from SAS Global Forum (SGF) and SAS Users Group International (SUGI) conferences; MWSUG, NESUG, PNWSUG, SCSUG, SESUG, and WUSS regional conferences; and PharmaSUG, PhUSE, and CDISC special-interest conferences.
CI-05 : Residential Energy Efficiency and the Principal-Agent Problem
Ryan Anderson, University of Nebraska Public Policy Center
Tuesday, 8:00 AM - 8:50 AM, Location: Senate Room B
Investments in residential energy-efficiency are associated with both positive externalities (reduced greenhouse gas emissions, reduced need for new power-generating capacity) and private cost savings. However, these investments lag far behind those levels predicted by conventional cost-benefit analysis. This paper explores one potential explanation for this efficiency gap, the principal-agent (PA) problem as it applies to rental housing. I employ SAS 9.3 to apply a logit model to data from the 2009 Residential Energy Consumption Survey and find that the PA problem is particularly pronounced in regards to weatherization improvements, a result that has implications for public policy. Somewhat advanced SAS techniques were used to report the results of this regression in terms of marginal effects. A basic familiarity with statistical processes and microeconomic theory is useful in reviewing this paper.
CI-06 : SAS® Treatments: One to One Marketing with a Customized Treatment Process
Dave Gribbin, SAS
Amy Glassman, SAS
Monday, 1:30 PM - 2:20 PM, Location: House Room A
The importance of sending the right message to the right person at the right time has never been more relevant than in today's cluttered marketing environment. SAS® Marketing Automation easily handles segment level messaging with out-of-the-box functionality. But how do you send the right message and the appropriately valued offers to the right person? And, how can an organization efficiently manage many treatment versions across distinct campaigns? This paper presents case studies of companies across different industries that send highly personalized communications and offers to their clientele using SAS® Marketing Automation. It describes how treatments are applied at a segment and a one-to-one level. It outlines a simple custom process that streamlines versioning treatments for reuse in multiple campaigns.
CI-07 : SAS PROC REPORT: How It Helps Improving College Student Retention Rate
Harjanto Djunaidi, Association of American Education Analytics
Monica Djunaidi, Association of American Education Analytics
Monday, 10:30 AM - 10:50 AM, Location: House Room A
This paper has shown how SAS PROC REPORT combined with other SAS PROCs can be utilized and applied by US colleges to improve their student retention and graduation rate. These education analytics approaches and tools are becoming more important after the Administration recently announced that College Affordability Rating (CAR) criteria will be used as the basis in awarding federal funding such as Pell grant to US Colleges. Higher learning institutions with higher CAR will be awarded more federal money compared to those who have lower rating. Time is running out for the colleges to improve their report card such as graduation rate which is one of the components measured in the CAR. These institutions have practically one academic year not only to improve the graduation rate, but also to reduce their tuition and their students' debt. Therefore, the ability to produce various reports and analyses timely that meet the decision makers' needs are important and no longer an option, but a must. Strategic decisions have to be made constantly, and they need to be supported by accurate data, professionally analyzed, provided and presented in a timely basis. Applying statistical approaches such as multivariate analyses, an Institutional Research Intelligence (IRI) expert is able to identify the odds of freshmen-year students to drop-out from their program. These estimation results combined with the output generated by PROC REPORT will help to group high risk student population so that appropriate policy and early intervention efforts can be done to minimize the possible damages. This new early alert approach applied to identify high risk students is increasingly vital in a more volatile education industry. It is even more vital, after the CAR regulation was announced. Except for the top-tier schools, most US colleges, either at a two-year or four-year public or private, not-for-profit or for-profit institutions are struggling to keep both under-prepared full-time and part-time, first-time students to survive their freshmen year courses. Therefore, they are more likely will face serious challenges to improve their graduation rate which in turns will reduce their ability to get funding from the federal government.
CI-08 : Big Data, Fast Processing Speeds
Gary Ciampa, SAS
Monday, 12:30 PM - 1:20 PM, Location: House Room A
As data sets continue to grow, it is important for programs to be written very efficiently to make sure no time is wasted processing data. This paper covers various techniques to speed up data processing time for very large data sets or databases, including PROC SQL, data step, indexes and SAS® macros. Some of these procedures may result in just a slight speed increase, but when you process 500 million records per day, even a 10 percent increase is very good. The paper includes actual time comparisons to demonstrate the speed increases using the new techniques.
CI-09 : Applying Customer Analytics to Promotion Decisions
Wanda Shive, SAS
Tuesday, 9:30 AM - 10:20 AM, Location: Senate Room B
For years, retailers have struggled to measure the effectiveness of their promotional advertising efforts. Harnessing the big data within their customer and transaction files continues to be a major challenge. Approaches for gleaning true customer insights from that data are becoming more common. Measuring total shopping behavior in conjunction with specific promotions provides a better understanding of the overall impact on profitability. This paper describes how retailers are utilizing customer analytics to measure the effect that mass promotions have on the total basket spend of customers and to identify the most relevant offers for each individual customer.
Data Visualization and Graphics
DV-01 : PROC SGPLOT over PROC GPLOTShruthi Amruthnath, Experis US, Inc.
Tuesday, 8:00 AM - 8:20 AM, Location: Senate Room A
SAS® offers different statistical graphic procedures for data visualization and presentation. SAS® 9.2 brought out new family of template-based graphical procedures to create high-quality graphics called Statistical Graphics (SG) procedures. These procedures are so powerful that, more complex information can be presented effectively with minimal coding. This paper will focus on employing SGPLOT versus GPLOT; applying SGPLOT to financial data to create reports for trending data, consolidated reports and yearly reports; and managing, displaying and styling procedural output using ODS PDF. The SGPLOT procedure has different types of graphical figures like bar charts, line graphs and scatter plots. This paper explains how to produce line graphs like line plot and area plot by setting different options. In this presentation, each of these topics is illustrated using different techniques with an example.
DV-03 : The Graph Template Language: Beyond the SAS/GRAPH® Procedures
Jesse Pratt, Cincinnati Children's Hospital Medical Center
Tuesday, 8:30 AM - 9:20 AM, Location: Senate Room A
The SGPLOT and SGPANEL procedures are powerful tools that are capable of producing many types of high quality graphs; however, these procedures have some limitations. What happens when one is asked to specifically produce a graph that these procedures cannot create? The Graph Template Language (GTL) is much more flexible when it comes to creating customized displays. This paper presents situations where the SGPLOT and SGPANEL procedures break down, then briefly introduces GTL, and finally uses GTL to generate the displays not possible in PROC SGPLOT and PROC SGPANEL.
DV-04 : SAS® Tile Charts: Thousands of Business Tips with One Click
Martha Hays, SAS Institute Inc
Tuesday, 9:30 AM - 11:20 AM, Location: Senate Room A
The ever growing complexity, increasing size and availability of business information makes getting to the critical issues a major challenge. A well organized SAS Tile Chart will grab and focus your attention on those critical issues that require you to take action. This seminar outlines the capabilities and business uses of the SAS Tile Chart. The Tile Chart makes very effective use of the color, size and data tips associated with multiple (drillable) tiles presented in a hierarchical grid to communicate business intelligence information for volumes of relevant data that is available for analysis.
DV-05 : Using ODS PDF, Style Templates, Inline Styles, and PROC REPORT with SAS® Macro Programs
Patrick Thornton, SRI International
Tuesday, 1:00 PM - 1:50 PM, Location: Senate Room B
A production system of SAS macro programs is described that modularize the generation of syntax to produce client-quality reports of descriptive and inferential results in a PDF document. The reusable system of macros include programs that save all current titles, footnotes, option settings, establish standard titles, footnotes and option settings, and initially create the PDF document. Macro programs may be called to generate PROC REPORT syntax to produce various tables of descriptive and inferential results with customized bookmarks, table numbers and footnotes. A custom style template is use to determine the look of the whole document, and a macro program of inline style definitions is used to define the defaults for each type of table. A version of the style macro program may be customized to accommodate the needs of each project, and inline style parameters maybe modified for any given macro call. A macro program to end the PDF creates a standard data documentation page, and restores all original titles, footnotes and option settings. This paper is designed for the intermediate to advanced SAS programmer using Foundation SAS Software on a Windows operating system.
DV-06 : Bordering on Success with PROC GMAP in SAS®: Utilizing Annotate Datasets to Enhance Your Maps
Kathryn Schurr, Spectrum Health
Jonathan Wiseman, Spectrum Health
Tuesday, 2:00 PM - 2:20 PM, Location: Senate Room B
PROC GMAP is a valuable tool for visualizing data. GMAP gives users the means to represent their data geographically so that audiences can relate to the data on multiple levels at once. One limitation of PROC GMAP that was identified previously is its inability to allow for multiple geographic regions (such as zip code areas, census tracts and counties) to be plotted all at once, especially if they do not have the shape of a polygon. This paper will discuss how PROC GMAP in conjunction with the %MAPLABEL and %ANNOTATE macros (1) creates a map in SAS® that annotates various locations on the map as well as providing an (x,y)-coordinate grid corresponding to the shapefile being used, (2) gives the user the capability to move labels to a more desirable location, and (3) draws borders of differing map regions regardless of shape.
DV-07 : A Concise Display of Multiple Response Items
Patrick Thornton, SRI International
Tuesday, 2:30 PM - 3:20 PM, Location: Senate Room B
Surveys often contain multiple response items, such as language where a respondent may indicate that she speaks more than one language. In this case, an indicator variable (1=Yes, 0=No) is often created for each language category. This paper shows how a concise tabulation of the count and percent of respondents with a Yes on one or more indicator variables may be obtained using PROC TABULATE and a MULTILABEL format. A series of indicator variables is used to create a binary variable and its base-10 equivalent, and a MULTILABEL format is created to properly aggregate observations with a Yes on two or more indicator variables.
DV-08 : Time Contour Plots
David Corliss, Magnify Analytic Solutions
Tuesday, 3:30 PM - 3:50 PM, Location: Senate Room B
A time contour plot is two-dimensional color chart for visualizing changes in a population or system over time. Data for one point in time appear as a thin horizontal band of color. Bands for successive periods are stacked up to make a two-dimensional, with the vertical direction showing changes over time. As a system evolves over time, different kinds of events have different characteristic patterns. Creation of Time Contour Plots is explained step by step. Examples are given in astrostatistics, biostatistics, econometrics and demographics.
Hands-On Workshops
HW-01 : Know Thy Data: Techniques for Data ExplorationAndrew Kuligowski, HSN
Charu Shankar, SAS Canada
Monday, 8:00 AM - 9:20 AM, Location: Judicial Room
Get to know the #1 rule for data specialists: Know thy data. Is it clean? What are the keys? Is it indexed? What about missing data, outliers, and so on? Failure to understand these aspects of your data will result in a flawed report, forecast, or model. In this hands-on workshop, You will learn multiple ways of looking at data and its characteristics. You will learn to leverage PROC MEANS and PROC FREQ to explore your data. You will learn how to use PROC CONTENTS and PROC DATASETS to explore attributes and determine whether indexing is a good idea. You will learn to employ powerful PROC SQL's dictionary tables to easily explore aspects of your data.
HW-02 : Fast Access Tricks for Large Sorted SAS Files
Russell Lavery, contractor
Monday, 9:30 AM - 10:50 AM, Location: Judicial Room
This HOW will provide an explanation of, and sample code for, several ways that you can, very quickly, find rows in sorted SAS files. The techniques are not generally known but can be useful when you need to retrieve specified rows from a large sorted file. This cartoon format of this presentation makes it a good review of the internals of the SAS data step and the program data vector. Even if you do not need these techniques, if you have never read these sections in the manuals this might be just the thing for you.
HW-03 : Getting Excel-lent Data Generating SAS Datasets from a Directory of Spreadsheets
Ben Cochran, The Bedford Group
Monday, 12:30 PM - 1:50 PM, Location: Judicial Room
Sometimes a SAS programmer has to convert many spreadsheets into many SAS datasets. This presentation starts with a directory full of spreadsheets. Next, a basic program is written to read the spreadsheets one at a time into a series of SAS datasets. Finally, the basic program is converted into a macro program that can dynamically read any number of spreadsheets into SAS datasets.
HW-04 : Using SAS® ODS Graphics
Chuck Kincaid, Experis Business Analytics
Tuesday, 8:00 AM - 9:20 AM, Location: Judicial Room
This presentation will teach the audience how to use SAS® ODS Graphics. Now part of Base SAS, ODS Graphics are a great way to easily create clear graphics that allow any user to tell their story well. SGPLOT and SGPANEL are two of the procedures that can be used to produce powerful graphics that used to require a lot of work. The core of the procedures are explained, as well as the options available. Furthermore, we explore the ways to combine the individual statements to make more complex graphics that tell the story better. Any user of Base SAS on any platform will find great value from the SAS ODS Graphics procedures.
HW-05 : Building a Better Bar Chart with SAS® Graph Template Language
Perry Watts, Stakana Analytics
Tuesday, 9:30 AM - 10:50 AM, Location: Judicial Room
This workshop combines instructions for chart building with principles defined by the statistical graphics experts: Edward Tufte, William Cleveland, and Naomi Robbins. Step-by-step instructions are provided for building basic and group charts that display frequencies, sums, percents, and means with associated confidence intervals. Group charts are further subdivided into repeating and non-repeating categories. For repeating group charts varying bar displays such as stacked, cluster, and nested will also be reviewed. Exercises make use of the BARCHART statement included in the SAS® 9.3 Graph Template Language Reference manual. Along the way, comparisons are made between GTL's BARCHART statement and the GCHART procedure in SAS/GRAPH software. The data used in the examples come either from the SASHELP library or from pre-summarized data read into WORK data sets with a DATALINES statement. While there may not be time in the workshop to cover advanced topics, two enhanced bar charts will be covered in the paper. Intermediate to advanced SAS programmers with experience using any graphics software package will get the most out of this presentation.
HW-06 : Hands-on SAS® Macro Programming Tips and Techniques
Kirk Paul Lafler, Software Intelligence Corporation
Tuesday, 1:00 PM - 2:20 PM, Location: Judicial Room
The SAS® Macro Language is a powerful tool for extending the capabilities of the SAS System. This hands-on workshop presents numerous tips and tricks related to the construction of effective macros through the demonstration of a collection of proven Macro Language coding techniques. Attendees learn how to process statements containing macros; replace text strings with macro variables; generate SAS code using macros; manipulate macro variable values with macro functions; handle global and local variables; construct arithmetic and logical expressions; interface the macro language with the DATA step and SQL procedure; store and reuse macros; troubleshoot and debug macros; and develop efficient and portable macro language code.
HW-07 : Using PROC FCMP to the Fullest: Getting Started and Doing More
Arthur Carpenter, CALOXY
Tuesday, 2:30 PM - 3:50 PM, Location: Judicial Room
The FCMP procedure is used to create user defined functions. Many users have yet to tackle this fairly new procedure, while others have only attempted to use only its simplest options. Like many tools within SAS®, the true value of this procedure is only appreciated after the user has started to learn and use it. The basics can quickly be mastered and this allows the user to move forward to explore some of the more interesting and powerful aspects of the FCMP procedure. Starting with the basics of the FCMP procedure, this paper also discusses how to store, retrieve, and use user defined compiled functions. Included is the use of these functions with the macro language as well as with user defined formats. The use of PROC FCMP should not be limited to the advanced SAS user; even those fairly new to SAS should be able to appreciate the value of user defined functions.
In-Conference Training
CT-01 : Top 10 SAS Best Programming Practices They Didn't Teach You in SchoolCharu Shankar, SAS Institute Canada
Tuesday, 8:00 AM - 11:20 AM, Location: Legislative Room (A, B)
How can programming time, I/O, CPU and memory be reduced? What is the data analysts' #1 rule? What are three questions you need to answer? What new features will help improve performance? These are age-old questions that have had data analysts thinking since the dawn of the first SAS program. Get all the answers in this new and informative seminar.
JMP
JM-01 : Dealing with Data below the Detection Limit: Limit Estimation and Data ModelingDiane Michelson, SAS Institute Inc.
Mark Bailey, SAS Institute Inc.
Monday, 8:00 AM - 8:50 AM, Location: Senate Room B
Measuring trace levels of contaminants in chemicals or gases can be difficult. When the signal is very small, it can be lost in the noise. Chemical analyses are characterized by their accuracy, precision, and linear range. A detection limit is the smallest amount the can be reliably detected by the procedure. The procedure can be used on a blank, with no amount of the substance, or on a sample containing the substance to be measured. Various methods of estimating a detection limit are compared. We will examine the differences between using tolerance intervals or confidence intervals on measurements on blanks, as well as regression approaches, including the use of linear regression by ordinary and weighted least squares. A contamination example will be demonstrated, using the Distribution platform, Fit Y by X, and Fit Model in JMP. Responses below the detection limit can be included in your data analysis. Ad hoc approaches produce biased estimates and should be avoided. Such responses are censored data, and likelihood methods exist to handle censoring. An example of a designed experiment is handled using the Parametric Survival personality in Fit Model in JMP.
JM-02 : 2x10-Minute JMP®
George Hurley, The Hershey Company
Monday, 9:30 AM - 9:50 AM, Location: Senate Room B
Heard of JMP®, but haven't had time to try it? Don't want to devote 50 minutes to a talk about software that you might not want to use? This is the talk to you. In the first 10 minutes, you will learn some of the amazing visualization and modeling features in JMP and how to use them. In the second 10 minutes, we'll go through a live example! This talk will JMP-start your JMP usage. When it's complete, we suspect you will want to attend some of the longer talks, too.
JM-03 : The JMP Journal: An Analyst's Best Friend
Nate Derby, Stakana Analytics
Monday, 9:00 AM - 9:20 AM, Location: Senate Room B
The JMP Journal is an incredibly useful tool for consultants and analysts, yet it's not commonly used. We first explain what the JMP Journal is, then describe how it can be effectively used to keep track of an analysis (and its underlying code), to present results to a boss or client, or to use as a collaboration tool.
JM-04 : Google® Search Tips and Techniques for SAS® and JMP® Users
Charles Edwin Shipp, Consider Consulting Corp
Kirk Paul Lafler, Software Intelligence Corp
Monday, 10:30 AM - 11:20 AM, Location: Senate Room B
Google® (www.google.com) is the world's most popular and widely-used search engine. As the premier search tool on the Internet today, SAS® and JMP® users frequently need to identify and locate SAS and JMP content wherever and in whatever form it resides. This paper provides insights into how Google works and illustrates numerous search tips and techniques for finding articles of interest, reference works, information tools, directories, PDFs, images, current news stories, user groups, and more to get search results quickly and easily.
JM-05 : Design of Experiments (DOE) using JMP
Charles Edwin Shipp, Consider Consulting Corp.
Monday, 1:30 PM - 2:20 PM, Location: Senate Room B
JMP has provided some of the best design of experiment software for years. The JMP team continues the tradition of providing state-of-the-art DOE support. In addition to the full range of classical and modern design of experiment approaches, JMP provides a template for Custom Design for specific requirements. The other choices include: Screening Design; Response Surface Design; Choice Design; Accelerated Life Test Design; Nonlinear Design; Space Filling Design; Full Factorial Design; Taguchi Arrays; Mixture Design; and Augmented Design. Further, sample size and power plots are available. We give an introduction to these methods followed by two examples with data. A lively discussion will follow.
JM-06 : When will it break? Exploring Product Reliability Using JMP.
Erich Gundlach, SAS
Monday, 12:30 PM - 1:20 PM, Location: Senate Room B
The reliability of your product strongly influences business success - whether you're making semiconductors, light bulbs, automobiles, shoes, medical devices or jet engines. Reliability analysis can help ensure that a product functions as intended throughout its life, guaranteeing happy customers who provide repeat business as well as referrals. Dependable products also reduce warranty costs, which can otherwise quickly eat away at your profit margins. Join Erich Gundlach and learn more about JMP's Reliability platform. Find out how you can model system reliability and increase mean time between failures.
JM-07 : JMP, and Teaching JMP: a panel discussion
Charles Edwin Shipp, Consider Consulting Corp
Monday, 10:00 AM - 10:20 AM, Location: Senate Room B
A panel of the invited speakers will each present a viewpoint on JMP software, JMP usage, and JMP user and management training. We will discuss the importance of management visibility, user group possibilities, and JMP training within your enterprise and in the community. You, the audience, will be an important part of the lively discussion that follows.
Pharmaceutical Apps
RX-01 : Introduction to REDCap for Clinical Data CollectionShannon Morrison, Cleveland Clinic
Monday, 8:00 AM - 8:50 AM, Location: House Room B
REDCap (Research Electronic Data Capture) is a free web-based tool that attempts to make collecting clinical data easy. The application is used primarily for creating databases and/or surveys by using the "Online Designer" in a web browser or constructing a "Data Dictionary" in Excel, which can be uploaded to REDCap. Multiple users can be added to a project and changes can be made quickly and easily online. Users have the ability to add validations and/or branching logic to variables, and once data collection is complete data can be automatically exported to common statistical packages...including SAS! This talk will focus on using the Online Designer to create databases in REDCap.
RX-02 : SAS and REDCap API: Efficient and Reproducible Data Import and Export
Sarah Worley, Cleveland Clinic
Dongsheng Yang, Cleveland Clinic
Monday, 9:00 AM - 9:20 AM, Location: House Room B
REDCap (Research Electronic Data Capture), a web application for the development and use of online research databases, allows SAS users to download automatically-generated SAS code for importing, labeling, and formatting data. REDCap Application Programming Interface (REDCap API) provides SAS users with the option to import and export data, files, and database settings through SAS programs. The use of REDCap API with both PC and server installations of SAS 9.3 will be demonstrated.
RX-03 : Let SAS® Do Your DIRty Work
Richann Watson, Experis
Monday, 9:30 AM - 9:50 AM, Location: House Room B
Making sure you have all the necessary information to replicate a deliverable saved can be a cumbersome task. You want to make sure that all the raw data sets are saved, all the derived data sets, whether they are SDTM or ADaM data sets, are saved and you prefer that the date/time stamps are preserved. Not only do you need the data sets, you also need to keep a copy of all programs that were used to produce the deliverable as well as the corresponding logs when the programs were executed. Any other information that was needed to produce the necessary outputs needs to be saved. All of these needs to be done for each deliverable and it can be easy to overlook a step or some key information. Most people do this process manually and it can be a time-consuming process, so why not let SAS do the work for you?
RX-04 : An Analysis of Risk Behavior Trends and Mental Health in American Youth Using PROC SURVEYLOGISTIC
Deanna Schreiber-Gregory, NDSU
Monday, 10:00 AM - 10:20 AM, Location: House Room B
The current study looks at recent mental health and risk behavior trends of youth in America. Data used in this analysis was provided by the Center for Disease Control and Prevention and gathered using the Youth Risk Behavior Surveillance System (YRBSS). Interactions between risk behaviors and reported mental states - such as depression, suicidal ideation, and disordered eating - are the main subject of this analysis. This study also outlines demographic differences in risk behaviors and mental health issues as well as correlations between the different mental health issues. A final regression model including the most significant contributing factors to suicidal ideation, depression, and disordered eating is provided and discussed. Results included reporting differences between the years of 1991 and 2011. All results are discussed in relation to current youth health trend issues. Data was analyzed using SAS® 9.3 and JMP® 10. This presentation is meant for any level of SAS User.
RX-05 : Reading and Writing RTF Documents as Data: Automatic Completion of CONSORT Flow Diagrams
Arthur Carpenter, CALOXY
Monday, 2:00 PM - 2:20 PM, Location: House Room B
Whenever the results of a randomized clinical trial are reported in scientific journals, the published paper must adhere to the CONSORT (CONsolidated Standards Of Reporting Trials) statement. The statement includes a flow diagram, and the generation of these CONSORT flow diagrams is always problematic, especially when the trial is not the typical two-arm parallel design. Templates of the typical two-arm design flow diagram are generally available as RTF documents, however the completion of the individual fields within the diagram is both time consuming and prone to error. The SAS Macro language was used to read a RTF template file for the CONSORT flow diagram of choice, fill in the fields using information available to the SAS program, and then rewrite the table as a completed RTF CONSORT flow diagram. This paper describes the process of reading and writing RTF files.
RX-06 : Creation and Implementation of an Analytic Dataset for a Multisite Surveillance Study using Electronic Health Records (EHR) and Medical Claims Data
Renuka Adibhatla, Healthpartners Institute for Education & Research
Gabriela VazquezBenitez, Healthpartners Institute for Education & Research
Mary Becker, Healthpartners Institute for Education & Research
Amy Butani, Healthpartners Institute for Education & Research
Monday, 10:30 AM - 10:50 AM, Location: House Room B
SAS code is utilized for research projects of the Health Maintenance Organization Research Network (HMORN). The HMORN is a consortium of 18 health care delivery organizations with integrated research divisions with over 15 million patients. Multisite research is conducted by utilizing the HMORN Virtual Data Warehouse (HMORN VDW), a distributed data network where each site locally stores their data in standardized data structures, in this case as SAS datasets. HMORN VDW is an excellent source of data for surveillance and observational studies of chronic conditions such as diabetes and cardiovascular disease.SAS programs are shared among HMORN project sites in order to extract site-specific data from each site which is then combined to represent the overall HMORN patient population and results. This paper focuses on the various programming techniques, including some borrowed from previous SAS conference papers and lessons learned from an ongoing HMORN multi-site project involving 11 HMORN research centers. These methods were applied to produce a clean analytic dataset with extensive use of PROC SQL, arrays, and macros. The paper clearly demonstrates the advantages of utilizing distributed SAS code for extracting data, the benefits from a HMORN VDW for multisite research, and the required organization of the SAS datasets and programs. The paper also discusses the challenges presented with data volumes and storage. SAS 9.2 was used on both Windows and UNIX environments. The paper is intended for beginner to intermediate level SAS programmers preferably with minimal knowledge of EHR and Claims data.
RX-07 : Examining Risk Factors of Referred and Substantiated Child Maltreatment in California Latino Infants Using PROC GENMOD
Kechen Zhao, University of Southern California
Monday, 11:00 AM - 11:20 AM, Location: House Room B
Count data are frequently encountered in social science, epidemiological research, and biomedical field. The GENMOD procedure is commonly used to build a Poisson regression model for count data. The use of PROC GENMOD is demonstrated in our study of California Latino children maltreatment. In this study, number of referred or substantiated child maltreatment incidences over a particular period is modeled by using Poisson regression. This study aimed to examine to what extent to which variations in referred and substantiated child maltreatment are attributable to the variations in child and maternal demographic and characteristics of California Latino population. Model building process by using PROC GENMOD alone with various features which were newly implemented in SAS® 9.2 will also be introduced in this paper.
RX-08 : A Tutorial on PROC LOGISTIC
Arthur Li, City of Hope
Monday, 12:30 PM - 1:20 PM, Location: House Room B
In the pharmaceutical or health care industries, we often encounter data with dichotomous outcomes, such as having (or not having) a certain disease. This type of data can be analyzed by building a logistic regression model via the LOGISTIC procedure. In this paper, we will address some of the model building issues that are related to logistic regression. In addition, some statements in PROC LOGISTIC that are new to SAS® 9.2 and ODS statistical graphics relating to logistic regression will also be introduced in this paper.
RX-09 : Using SAS in Clinical Programming Project Management from Excel Spreadsheets
Jean Crain, InVentiv Health Clinical, LLC
Monday, 1:30 PM - 1:50 PM, Location: House Room B
Tracking Clinical Programming Project Management as we do from Microsoft Excel© Spreadsheets can give you a wealth of information about your projects if you have standardized format and method of entry by programming and statistics. From our Excel© spreadsheets, using Base SAS©, SAS Macros and Proc Report we can retrieve the following data and present valuable information on how long a time period project spans, number of resources, time spent, productivity, and audit checking. The data obtained can provide multiple reports for analysis: Number of Programs Number of Outputs Number of Programs by Type (Analysis Dataset, Graph, Listing, Summary) Number of Outputs by Type (Analysis Dataset, Graph, Listing, Summary) Number of Programs by Employee (Programming, Validation and Statistical Validation and Review) Number of Outputs/Reviews by Employee Percent completion on projects by type: Original Programming, Validation Programming, Initial Statistical Review, Senior Statistical Review (number of outputs completed / number of outputs). First program completion date Last program completion date Number of calendar days (last date - first date), Overall and by Type. Number of work days (count unique dates), Overall and by Type. Audit check: Do Initials for Programming and Statistical Activities have a matching Name on Signature Page? Comment Data to analyze for trends: When programs/outputs need correction, does that indicate there are there areas for improvement in Requirements? Do the same issues come up from project to project? We can use this because we do have a standard for the spreadsheet tracking and entries by programming and statistics staff. Our hours are tracked elsewhere, but this will give you a good accounting of number of days for projections and it is an informative method of conducting process improvement. This is not complex programming but it demonstrates many functions in Base SAS, Proc Import of Excel spreadsheet data, and Proc Report to create multiple reports.
Posters
PT-01 : Exploring the PROC SQL _METHOD OptionKirk Paul Lafler, Software Intelligence Corporation
(Meet the Poster Authors session, Monday, 2:30 PM - 3:00 PM, Legislative Foyer)
The SQL Procedure contains powerful options for users to take advantage of. This poster illustrates the fully supported _METHOD option as an applications development and tuning tool. Attendees learn how to use this powerful option to better understand and control how a query processes.
PT-02 : Macro to Compute Best Transform Variable for the Model
Nancy Hu, ASA
(Meet the Poster Authors session, Monday, 2:30 PM - 3:00 PM, Legislative Foyer)
This study is intended to assist Analysts to generate the best of variables using simple arithmetic operators (square root, log, loglog, exp and rcp) and such as monthly amount paid, daily number of received customer service calls, weekly worked hours on a project, or annual number total sales for a specific product. During a statistical data modeling process, Analysts are often confronted with the task of computing derived variables using the existing available variables. The advantage of this methodology is that the new variables may be more significant than the original ones. This paper gives a new way to compute all the possible variables using a set of math transformation. The codes include many SAS features that are very useful tools for SAS programmers to incorporate in their future codes such as %SYSFUNC, SQL, %INCLUDE, CALL SYMPUT, %MACRO, DICTIONARY.XXXX (where XXXX can be TABLE, COLUMN), SORT, CONTENTS, MERGE, MACRO _NULL_, as well as %DO & %TO & and many more.
PT-03 : You Could Be a SAS® Nerd If . . .
Kirk Paul Lafler, Software Intelligence Corporation
(Meet the Poster Authors session, Monday, 2:30 PM - 3:00 PM, Legislative Foyer)
Are you a SAS® nerd? The Wiktionary (a wiki-based Open Content dictionary) definition of nerd is a person who has good technical or scientific skills, but is generally introspective or introverted. Another definition is a person who is intelligent but socially and physically awkward. Obviously there are many other definitions for nerd , many of which are associated with derogatory terms or stereotypes. This presentation intentionally focuses not on the negative descriptions, but on the positive aspects and traits many SAS users possess. So let's see how nerdy you actually are using the mostly unscientific, but fun, Nerd detector.
PT-04 : Trend Reporting Using the MXG® Trend Performance Database
Neal Musitano Jr., U.S. Department of Veterans Affairs
(Meet the Poster Authors session, Monday, 2:30 PM - 3:00 PM, Legislative Foyer)
This paper is about my user experience on trend reporting for z/OS server performance using the MXG Trend Performance Database (PDB). In order to implement trend reporting with graphs and charts, the trend pdb must be built first. Thus my user experience of building the trend pdb is included. MXG users build a daily pdb and they use it to produce daily reports for z/OS performance monitoring. However; because of daily tasks or firefighting duties, or deadlines in getting out daily reports or other reasons they sometimes overlook building the trend performance database. Additional reasons for the delaying the trend pdb implementation, include uncertainties of what is involved in building the trend pdb, what to include in it, unfamiliarity of the trend pdb structure and what information to collect for the trends. This paper will address those issues to help users get started with z/OS trend reporting and building the MXG trend performance database.
PT-05 : Customizing a Multi-Cell Graph Created with SAS ODS Graphics Designer
Yanhong Liu, Cincinnati Children's Hospital Medical Center
(Meet the Poster Authors session, Monday, 2:30 PM - 3:00 PM, Legislative Foyer)
Combining multiple graphs and/or statistical data tables into one graph is an effective way to compare research data side by side or to present a summarized report of related information together. The SAS® Graph Template Language (GTL) provides the ability to create such multi-cell graphs by using its powerful building-block syntax. In SAS® 9.3, GTL provides a new feature, discrete attribute maps , which enables us to map visual attributes such as shape and color to input data values. You can add this statement block to the GTL code and customize the appearance of each cell of the multi-cell graph, thereby improving the overall graphical visualization, and allowing better interpretation of the graph. To avoid writing GTL code from scratch, we can use the ODS Graphics Designer which is based on Graph Template Language to create the graph, then copy the GTL code generated by the ODS Graphics Designer into the program editor for further customization.
PT-06 : Aligning Parallel Axes in SAS® GTL
Perry Watts, Stakana Analytics
(Meet the Poster Authors session, Monday, 2:30 PM - 3:00 PM, Legislative Foyer)
Sometimes it is important to display data using parallel axes that need to be aligned. Unfortunately Graph Template Language (GTL) does not have an ALIGN=TRUE|FALSE axis option. Therefore to compensate, GTL options are manipulated so that weights can be displayed in pounds and kilograms, heights in inches and centimeters, frequencies in counts and percents, and temperatures in Celsius and Fahrenheit. The following GTL statements are used in version 9.3 SAS to create the graphs: SCATTERPLOT, HISTOGRAM, BARCHART, and VECTORPLOT. Intermediate to advanced SAS programmers with experience using any graphics software package will get the most out of this presentation.
PT-07 : Increase Pattern Detection in SAS® GTL with New Categorical Histograms and Color Coded Asymmetric Violin Plots
Perry Watts, Stakana Analytics
(Meet the Poster Authors session, Monday, 2:30 PM - 3:00 PM, Legislative Foyer)
Detecting patterns in graphics output is much easier when numeric data can be grouped categorically. Such is the case with the Body Mass Index and its four classifications: underweight, normal weight, overweight and obese. This presentation goes from conventional histogram to asymmetric violin plot with coverage of the categorical histogram along the way. HISTOGRAM, BANDPLOT and LATTICE statements are described in context. Version 9.3 SAS must be used to replicate the graphs. Intermediate to advanced SAS programmers with experience using any graphics software will get the most out of this presentation.
PT-08 : The DO's and DON'Ts of PROC REPORT: Building From Introductory Ideas into Professional Results
Daniel Sturgeon, Priority Health
Erica Goodrich, Grand Valley State University/MPI Research
(Meet the Poster Authors session, Monday, 2:30 PM - 3:00 PM, Legislative Foyer)
It is important to be able to effectively display results in a way that is both useful and aesthetically pleasing. PROC REPORT's flexibility to procedure reports in SAS teamed with PROC TEMPLATE and ODS graphics and outputs can create professional looking reports. However for a user who is not well versed in using SAS or PROC REPORT this coding can seem daunting and downright foreign since the syntax does not always match other SAS procedures. In this paper we will address the do's and don'ts of PROC REPORT touching on methods, options, and tricks that can be used; while discussing some of the more common mistakes users make that prevent them from creating the best looking reports possible.
PT-09 : Twin Ports Area SAS Users Group: Doing great things on a great lake
Steve Waring, Essentia Institute of Rural Health (EIRH)
Ron Regal, University of Minnesota-Duluth/EIRH
Paul Hitz, EIRH
(Meet the Poster Authors session, Monday, 2:30 PM - 3:00 PM, Legislative Foyer)
The Twin Ports Area SAS Users is a newly formed local user group comprised of SAS users from academia, health research, and business in and around the Duluth (MN) -Superior (WI) area. Our mission is to promote all things SAS among current users and the SAS-curious, as well as foster teaching, research, and technical development collaborations that take advantage of the many strengths of SAS. We held our first meeting in November 2012, became an officially recognized SAS local user group and launched our website in January, and have already hosted a seminar led by a SAS speaker. We meet bimonthly in a casual forum to discuss and share topics ranging from specific programming issues to better utilization of SAS software. This presentation is for any skill level with the intent of providing the audience with an introduction of our group, highlighting some of the areas of teaching and research our various members are engaged in and our plans for the future.
PT-10 : SAS Enterprise Guide® - Implementation Hints and Techniques for Insuring Success With Traditional SAS Programmers
Roger Muller, Data-to-Events.com
(Meet the Poster Authors session, Monday, 2:30 PM - 3:00 PM, Legislative Foyer)
Traditional SAS programmers develop SAS code in files and submit it for processing regardless of the operating environment (PC, Unix, etc.). SAS Enterprise Guide (EG) follows this model, but adds some unique additional capabilities. This paper addresses setup, initialization and workflow ideas to smoothen and enhance the transition to the EG graphical user interface centered around a process flow window. Work flow will be addressed in either standalone PC mode, or in conjunction with other servers (local vs. remote processing). Starting with hardware and network capabilities, the paper then moves into a discussion on data location and how that affects work flow. Dual screen systems, split views and internal vs. external SAS code storage will be addressed. Certain features in SAS EG may either be hidden or exposed with advantages to doing either. The number of process flows in an Enterprise Guide Project offers flexibility in constructing the total programming effort. Techniques for submitting developing code for step-by-step processing vs. the submission of entire project files is discussed. The paper will address environmental settings for both EG itself and for the advanced code editor. And lastly, the handiest key in EG, the F4 key, which is used to toggle back-and-forth between the current Process Flow window and the most recent window (program, data set, log, output, etc.) will be repeatedly emphasized. The importance of understanding a strong file backup process relative to the work flow being followed will also be discussed.
PT-11 : Transposing a Dataset from 'Long' to 'Wide'
Peter Batra, Univ. of Michigan
(Meet the Poster Authors session, Monday, 2:30 PM - 3:00 PM, Legislative Foyer)
This presentation will demonstrate how to transpose a long' dataset into a wide' dataset. In the example, we start with a dataset that contains a list of up to 10 drugs assigned to a person ID. In this dataset each record contains the data by drug. The end result is a dataset where each record (observation) contains the combined information from all drugs for by a person ID. Additionally, I show how to add information on the original long' file for each drug and then combine the wide' dataset with an existing ID level dataset. The end result is a very large dataset that contains all of the data at the person ID level. The presentation will show all of the code used to achieve this conversion from a 'long' to a 'wide' file. The following PROC's and the Data step were used to achieve this goal: CONTENTS, SORT, SQL, TRANSPOSE. The SAS code used in this presentation was run on SAS for Windows (9.2) on a 64 bit platform. This presentation is suitable for someone who has some familiarity with SAS coding and wants to add to their knowledge base with the steps involved to transpose a dataset from long' to wide'.
Rapid Fire
RF-01 : Getting Wild with ImportsKathryn Schurr, Spectrum Health Corporate
Jonathan Wiseman, Spectrum Health Corporate
Tuesday, 1:30 PM - 1:40 PM, Location: House Room B
ABSTRACT Ever have many similar datasets to import and not enough time to come up with a good macro to import and merge them all together? Using a wildcard import in SAS® v9.3 will solve all of your woes. This paper presents an efficient technique for importing multiple data files under one folder at the same time. Then it will further address the wildcard import as a SAS® Macro that can be used time and again for various folders containing differing data. This is a great tool for records kept electronically per facility, per student, or per patient; and especially when an extravagant query based data warehouse is unavailable.
RF-02 : Does foo Pass-Through? SQL Coding Methods and Examples using SAS® software
Stephen Crosbie, UM (Universal McCann)
Tuesday, 1:45 PM - 1:55 PM, Location: House Room B
Rapid Fire: Processing data between SAS® software and a DBMS, such as Netezza® or DB2®. Application of the SQL Procedure with the CONNECT Statement (requires SAS/CONNECT® software) and use of the PROC SQL Pass-Through Facility. While a SAS library reference gives PROC SQL access to DBMS tables (requires SAS/ACCESS® software), using Pass-Through SQL coding instead can often provide faster results. This paper provides useful examples of queries that were passed-through to a DBMS to take advantage of greater processing resources or better query optimization.
RF-04 : Copy and Paste from Excel to SAS®
Arthur Tabachneck, myQNA, Inc.
Matthew Kastin, I-Behavior, Inc.
Tuesday, 1:00 PM - 1:10 PM, Location: House Room B
Most, if not all, of us are far too familiar with the problems that one can confront when trying to import an Excel workbook into a SAS® dataset. Numeric fields might be imported as character fields, dates might be imported as either character fields or dates that are sixty (60) years earlier than what they actually represent, and fields containing time values might be imported as representing one-half of a second when they actually represent 43,200 seconds. While such discrepancies can be corrected by following the use of PROC IMPORT with a carefully written datastep, the present paper presents an alternative that uses less code, only requires base SAS and, on our test data, ran almost fifty (50) times faster than using PROC IMPORT.
RF-05 : Increase Your Productivity by Doing Less
Arthur Tabachneck, myQNA, Inc.
Xia Ke Shan, Chinese Financial Electrical Company
Robert Virgile, Robert Virgile Associates, Inc.
Joe Whitehurst, High Impact Technologies
Tuesday, 1:15 PM - 1:25 PM, Location: House Room B
Using a keep dataset option when declaring a data option has mixed results with various SAS® procedures. It might have no observable effect when running PROC MEANS or PROC FREQ but, if your datasets have many variables, it could drastically reduce the time required to run some procs like PROC SORT and PROC TRANSPOSE. This paper describes a fairly simple macro that could easily be modified to use with any proc that defines which variables should be kept and, as a result, make your programs run 12 to 15 times faster.
RF-06 : Reuse, Don't Reinvent: Extending Model Selection Using Recursive Macro
Anca Tilea, University of Michigan
Philip Francis III, Eastern Michigan University
Tuesday, 2:15 PM - 2:25 PM, Location: House Room B
The existing SAS® macro: %model_select - used to recursively select the best' model based on the R2 selection method - is performing great for specific data and specific variables. The steps of the macro %model_select are as follows: perform a BEST SUBSETS selection based on R2 to get the list of candidate models, calculate the estimates and associated p-values, eliminate the covariates that are never significant, and repeat. This macro does not allow for multilevel variables to be considered for model selection, it does not allow for hierarchical modeling, and it only considers one model selection method - R2. This paper aims to enhance existing SAS® methods, specifically, by allowing various model selection methods (e.g., adjusted-R2, Mallow's Cp, etc.) and the inclusion of categorical variables with multiple levels. The added capabilities will allow the macro to still be user friendly, yet be more robust. Several PROCEDURES, CALL SYMPUT, SCAN, DO-LOOP, IF-THEN-ELSE statements and functions, and various Output Delivery System (ODS) statements are used in expanding the %model_select macro. This paper is intended for the intermediate SAS® user, with intermediate statistical skills and SAS® 9.1 or above.
RF-07 : Data Review Information: N-Levels or Cardinality Ratio
Ronald Fehd, SAS-L
Tuesday, 2:30 PM - 2:40 PM, Location: House Room B
This paper reviews the database concept: Cardinality Ratio. The SAS(R) frequency procedure can produce an output data set with a list of the values of a variable. The number of observations of that data set is called N-Levels. The quotient of N-Levels divided by the number-of-observations of the data is the variable's Cardinality Ratio (CR). Its range is in (0--1] Cardinality Ratio provides an important value during data review. Four groups of values are examined.
RF-08 : Improve Your ODS Experience with These Essential VBScript Tools
Rose Grandy, Abbott Laboratories
Tuesday, 2:45 PM - 2:55 PM, Location: House Room B
While so much is available in version 9.2 and 9.3 of SAS® to make practically perfect submission-ready output using the Output Delivery System (ODS), there are still a few important things that just cannot be done with tools available within SAS. This paper will present some very useful SAS macros that will write and execute VBScript code to: convert ODS RTF files to true DOC files; merge table cells in ODS RTF files; convert RTF or DOC files to PDF; and, convert XML files created using ODS TAGSETS.EXCELXP to true Excel files.
RF-09 : An Efficient Approach to Automatically Convert Multiple Text Files (.TXT) to Rich Text Format Files (.RTF) Using SAS
Xingxing Wu, inVentiv Health
Jyoti Rayamajhi, Eli Lilly and Company
Tuesday, 3:00 PM - 3:10 PM, Location: House Room B
A large number of legacy and Third-Party-Organization (TPO) provided clinical trial statistical analysis output files stored in SAS Drug Development (SDD) are in .txt (text) format. Because of many advantages of rich text format (RTF) files compared with text files, it's a common task to convert these files from text format to RTF to meet the submissions, regulatory responses, or other requirements. In addition, clinical reviewers often desire to have the RTF outputs since they can write comments on them. This paper provides an efficient and easy-to-use approach to automatically convert multiple text files to RTF files. This approach can directly run in SDD without relying on other tools, such as Microsoft Office VBA, and SDD desktop connection tool. Furthermore, it can also be directly used in other SAS development environment, such as PC SAS. In this paper, an innovative approach is also proposed to resolve the issue that the underline character _ in the text files cannot be displayed after converted to RTF files in some situations. Compared with other approaches, the proposed one has the advantage of robustness. It can be directly applied to any situation without requiring the users to adjust the code to fit into their own situations. This paper is intended for the audiences with some general knowledge about Base SAS and RTF.
RF-10 : Maintaining Formats when Exporting Data from SAS into Microsoft Excel
Nate Derby, Stakana Analytics
Colleen McGahan, BC Cancer Agency
Tuesday, 3:15 PM - 3:25 PM, Location: House Room B
Data formats often get lost when exporting from SAS into Excel using common procedures such as PROC EXPORT or the ExcelXP tagset. In this paper we describe some tricks to retain those formats.
RF-11 : Don't Get Blindsided by PROC COMPARE
Joshua Horstman, Nested Loop Consulting
Roger Muller, Data-to-Events.com
Tuesday, 3:30 PM - 3:40 PM, Location: House Room B
"NOTE: No unequal values were found. All values compared are exactly equal." That message is the holy grail for the programmer tasked with independently replicating a production dataset to ensure its correctness. Such a validation effort typically culminates in a call to PROC COMPARE to ascertain whether the production dataset matches the replicated one. It is often assumed that this message means the job is done. Unfortunately, it is not so simple. The unwary programmer may later discover that significant discrepancies slipped through. This paper surveys some common pitfalls in the use of PROC COMPARE and explains how to avoid them.
RF-12 : The Utility of the DATA Step Debugger in Logic Errors
Darryl Nousome, University of Southern California
Tuesday, 2:00 PM - 2:10 PM, Location: House Room B
There are a few strategies in debugging errors that arise during SAS® programming. Errors in programming can manifest as either syntax or logic errors. Syntax errors tend to be easily identified as they stop the program and generate error messages in the log window. While logic errors will not halt SAS during the DATA step compilation phase, this may result in data that is unintended. A useful tool in examining logic errors is invoking the DATA Step Debugger, which allows viewing of the Program Data Vector (PDV) during the execution of the DATA step. When the DATA Step Debugger is running, there are many options that can examine values of selected variables, suspend statements, display the values of any variables in the PDV, or other commands. This process can aid in identifying segments of code that may contain logic errors. This paper will highlight some techniques that are used by the DATA Step Debugger.
SAS 101
S1-01 : The Essence of DATA Step ProgrammingArthur Li, City of Hope National Medical Center
Monday, 8:00 AM - 8:50 AM, Location: Legislative Room A
The fundamental of SAS® programming is DATA step programming. The essence of DATA step programming is to understand how SAS processes the data during the compilation and execution phases. In this paper, you will be exposed to what happens behind the scenes while creating a SAS dataset. You will learn how a new dataset is created, one observation at a time, from either a raw text file or an existing SAS dataset, to the program data vector (PDV) and from the PDV to the newly-created SAS dataset. Once you fully understand DATA step processing, learning the SUM and RETAIN statements will become easier to grasp. Relating to this topic, this paper will also cover BY-group processing.
S1-02 : The Power of Macros
Audrey Yeo, Aviva USA
Monday, 9:00 AM - 9:20 AM, Location: Legislative Room A
The SAS® Macro facility is an extremely powerful tool that should be in the toolbox of every SAS programmer. However, without some proper training it is difficult to implement, or when it is implemented it often results in hard to understand code. On the other hand once you have mastered the macro facility, it opens up a whole new world. This paper will show some examples of creating using SAS macros to help simplify coding and reduce repetition or duplication of code.
S1-03 : Data Cleaning 101: An Analyst's Perspective
Anca Tilea, University of Michigan
Deanna Chyn, University of Michigan
Monday, 1:30 PM - 1:50 PM, Location: Legislative Room A
On a daily basis, we are faced with data, both clean and dirty. SAS" offers a multitude of ways to clean and maintain data. It is up to us, the analysts, to choose the best way. Often times the choice we make depends on the analysis needed further down the road. When you start as a novice analyst you are familiar with the basics of data steps and procedures: SET statement, PROC MEANS, PROC FREQ, PROC SORT. We hope this paper will provide insight to a SAS" user that has this basic knowledge but wants to code more efficiently. We will introduce you to some basic SAS" procedures (PROC TRANSPOSE, PROC SQL) and SAS" tricks (using CALL SYMPUT, DO-LOOP, IF-ELSE, and PRX-functions) that are beyond a simple data step, but not too complicated to be understood by someone with basic SAS" skills. This paper is intended for the novice SAS user, with basic to intermediate skills and SAS 9.1 or above.
S1-04 : Using SAS® to Analyze Data Submitted to the National Healthcare Safety Network (NHSN)
Michelle Hopkins, Stratis Health
Monday, 10:00 AM - 10:20 AM, Location: Legislative Room A
Hospitals across the nation are required by the Centers for Medicare & Medicaid Services (CMS) to submit data on a variety of Healthcare-associated infections (HAIs). These are infections that are acquired in the course of receiving care for other health conditions. These HAIs include Central Line-associated Bloodstream Infection, Catheter-associated Urinary Tract Infections, Surgical Site Infections, Clostridium difficile Infection, etc. The number of HAIs in the United States is alarming, costly, and can be deadly. Hospitals report this data to the Center for Disease Control's National Healthcare Safety Network (NHSN). Using SAS® 9.2, this paper will show how you can use the data found on NHSN for analysis. This analysis includes importing data, summarizing infection rates, and using the data to drive much needed improvement.
S1-05 : Strategies and Techniques for Debugging SAS® Program Errors and Warnings
Kirk Paul Lafler, Software Intelligence Corporation
Monday, 9:30 AM - 9:50 AM, Location: Legislative Room A
As a SAS® user, you've probably experienced first-hand more than your share of program code bugs, and realize that debugging SAS program errors and warnings can, at times, be a daunting task. This presentation explores the world of SAS errors and warnings, provides important information about syntax errors, input and output data sources, system-related default specifications, and logic scenarios specified in program code. Attendees learn how to apply effective techniques to better understand, identify, and fix errors and warnings, enabling program code to work as intended.
S1-06 : Power Trip: A Road Map of PROC POWER
Melissa Plets, Grand Valley State University
Julie Strominger, Grand Valley State University
Monday, 2:00 PM - 2:50 PM, Location: Legislative Room A
Sample size and power analyses are extremely important components to consider when designing, planning and recruiting for prospective clinical research projects. Unfortunately, these are often considered to be scary and complicated calculations for research clinicians. PROC POWER offers a systematic solution to finding the balance between efficiency and conclusive results, while remaining simple enough for non-statisticians to use. More power is universally considered to be advantageous, even outside of the domain of statistics. It is especially important in statistics, as more power results in a higher probability that the null hypothesis will be rejected when it is false. Additionally, sample size is directly related to power. In general, a larger sample size will result in more accuracy, precision, and higher power. An important aspect of study design is maximizing power while remaining within the bounds of financial feasibility. This is where PROC POWER comes in, providing methods of determining sample sizes and power calculations. Throughout this paper, we will construct a road map of the POWER procedure to assist both statisticians and non-statisticians with the implementation of this POWERful SAS tool.
S1-07 : Insurance Designation Randomization Application or How to Automate Your Bragging
Irvin Snider, Assurant Health
Monday, 12:30 PM - 12:50 PM, Location: Legislative Room A
Having taken over 65 insurance courses over the course of a 25 year home office career, I have acquired 15 insurance designations to list behind my name. This causes problems because they usually never fit neatly on one line even if only the initials are used and to some, it appears ostentatious. To solve this problem, I have developed SAS code to randomly select only five of the designations at any one time to place behind my name in my emails. Thus the list appears to be fresh with each email and I do not appear to be bragging (well, not too much). This presentation will include the topics of random selection, placing this output in a macro variable via CALL SYMPUT and calling the macro variable in the email message.
S1-08 : Let the CAT Out of the Bag: String Concatenation in SAS 9
Joshua Horstman, Nested Loop Consulting
Monday, 1:00 PM - 1:20 PM, Location: Legislative Room A
Are you still using TRIM, LEFT, and vertical bar operators to concatenate strings? It's time to modernize and streamline that clumsy code by using the string concatenation functions introduced in SAS 9. This paper is an overview of the CAT, CATS, CATT, and CATX functions introduced in SAS 9.0, and the new CATQ function added in version 9.2. In addition to making your code more compact and readable, this family of functions also offers some new tricks for accomplishing previously cumbersome tasks.
S1-09 : Introducing a Colorful PROC TABULATE
Ben Cochran, The Bedford Group
Monday, 10:30 AM - 11:20 AM, Location: Legislative Room A
Several years ago, one of my clients was in the business of selling reports to hospitals. He used PROC TABULATE to generate part of these reports. He loved the way this procedure 'crunched the numbers', but not the way the final reports looked. He said he would go broke if he had to sell naked PROC TABULATE output. So, he wrote his own routine to take TABULATE output and render it through Crystal Reports. That was before SAS came out with the Output Delivery System (ODS). Once he got his hands on SAS ODS, he kissed Crystal Reports good-bye. This paper is all about using PROC TABULATE to generate fantastic reports. If you want to generate BIG money reports with PROC TABULATE, this presentation is for you.
S1-10 : Effectively Utilizing Loops and Arrays in the DATA Step
Arthur Li, City of Hope National Medical Center
Tuesday, 8:00 AM - 8:50 AM, Location: Governor's Ballroom (C, D, E)
The implicit loop refers to the DATA step repetitively reading data and creating observations, one at a time. The explicit loop, which utilizes the iterative DO, DO WHILE, or DO UNTIL statements, is used to repetitively execute certain SAS® statements within each iteration of the DATA step execution. Utilizing explicit loops is often used to simulate data and to perform a certain computation repetitively. However, when an explicit loop is used along with array processing, the applications are extended widely, which includes transposing data, performing computations across variables, etc. Being able to write a successful program that uses loops and arrays, one needs to know the contents in the program data vector (PDV) during the DATA step execution, which is the fundamental concept of DATA step programming. This paper will cover the basic concepts of the PDV, which is often ignored by novice programmers, and then will illustrate how utilizing loops and arrays to transform lengthy code into more efficient programs.
S1-11 : SAS: Tips and Tricks
Audrey Yeo, Aviva USA
Tuesday, 9:00 AM - 9:20 AM, Location: Governor's Ballroom (C, D, E)
Using SAS® in the work environment is different from using SAS in the educational environment. Data in the work environment is neither as perfect nor as simple as data in the educational environment. This paper will highlight some tips and tricks that will help a new SAS user, whether a recent graduate or just new to SAS, deal with some common but challenging problems.
S1-12 : Data Presentation 101: An Analyst's Perspective
Anca Tilea, University of Michigan
Deanna Chyn, University of Michigan
Tuesday, 9:30 AM - 9:50 AM, Location: Governor's Ballroom (C, D, E)
You are done with the tedious task of data cleaning, and now the fun begins. You have several results and must present them in a useful manner to a statistician (if you are lucky) or a non-statistician (as is usually the case). SAS" provides a multitude of resources to do this very thing: HISTOGRAM statement in PROC UNIVARIATE and/or PROC SGPLOT, PROC REPORT, Output Delivery System statements for various PROCs, and so on. This paper will describe a few of these methods that we, as data analysts, have found to be most helpful when presenting results. This paper is intended for the novice SAS user, with basic to intermediate skills and SAS" 9.2 or above.
S1-13 : I Heart SAS Users
Joanne Ellwood, Progressive Insurance
Tuesday, 10:00 AM - 10:20 AM, Location: Governor's Ballroom (C, D, E)
In my 20 some years experience, I have found SAS users to be smart, helpful and highly motivated. I would like to share some of the experiences I have had in supporting SAS users I am very humbled whenever I find a SAS user snagged in a simple error and am able to be of assistance in helping to resolve the error. It is an honor and a privilege to work with such magnificent people as the much heartd SAS users. This paper is targeted for anyone who has ever used SAS.
S1-14 : A Poor/Rich SAS® User's PROC EXPORT
Arthur Tabachneck, myQNA, Inc.
Tom Abernathy, Pfizer, Inc.
Randy Herbison, Westat
Matthew Kastin, I-Behavior, Inc.
Tuesday, 10:30 AM - 10:50 AM, Location: Governor's Ballroom (C, D, E)
Have you ever wished that with one click you could copy any SAS® dataset, including variable names, so that you could paste the text into a Word file, powerpoint or spreadsheet? You can and, with just base SAS, there are some little known but easy to use methods that are available for automating many of your (or your users') common tasks.
S1-15 : Writing Macro Do Loops with Dates from Then to When
Ronald Fehd, SAS-L
Tuesday, 11:00 AM - 11:20 AM, Location: Governor's Ballroom (C, D, E)
Dates are handled as numbers with formats in SAS(R) software. The SAS macro language is a text-handling language. Macro \%do statements require integers for their start and stop values. This article examines the issues of converting dates into integers for use in macro \%do loops. Two macros are provided: a template to modify for reports and a generic calling macro function which contains a macro \%do loop that can return items in the associative array of dates. Example programs are provided which illustrate unit testing and calculations to produce reports for simple and complex date intervals.