HOME
QUICK SEARCH
AUTHORS
KEYWORDS
HELP
BUGSSGF 2021 Papers - Index
90 papers. Source: SAS Global Forum conference proceedings.
| # | Title× | Authors× | Abstract× | Keywords× | Extras | Pages |
|---|---|---|---|---|---|---|
| 1 | A Framework for Simple and Efficient Bootstrap Validation in SAS, with Examples | Isaiah Lankham, Matthew T. Slaughter |
Validation is essential for assessing a predictive model's performance with respect to optimism or overfitting. While traditional sample-splitting techniques like cross validation require us to divide our data between model building and model assessment, bootstrap validation enables us to use the full sample for both. This paper demonstrates a simple method for efficiently calculating bootstrap-corrected measures of predictive model performance in SAS. While several SAS procedures have options for automatic cross validation, bootstrap validation requires a more manual process. Examples focus on logistic regression using the LOGISTIC procedure in SAS/STAT, with additional discu ssion of how these techniques can be extended to other procedures and statistical models. The running example can be downloaded in Jupyter Notebook and .sas file format from https://github.com/saspy-bffs/sgf-2021-bootstrap-validation |
bootstrap, sample, data cleaning, proc step, proc logistic, logistic distribution, ods output, output, feature selection, key performance indicator, proc hplogistic, proc phreg, regression analysis, proc surveyselect, sample size, correlation, macro variable, proc plm, training set, by group processing, logistic regression, proc sql | source | 22 |
| 2 | A SAS Macro for Dynamic Page Breaks | Manohar Modem, Bhavana Bommisetty |
In clinical domain, we usually create many safety and efficacy tables with various statistics. While creating these tables, we introduce SAS dataset with statistics into PROC REPORT to create listing or rtf output. Using PROC REPORT-BREAK-PAGE, we can make sure that each parameter statistics starts in a new page in the output. If we want to make sure that a group of statistics does not break abruptly between pages, we may need to use conditional statements to assign page numbers. Whenever there is an update in mock shell or data, the number of rows in the output may increase or decrease, which in turn requires an update in conditional statements to prevent abrupt, breaks in the output. This led to an effort to create a macro, which prevents abrupt page breaks and provides meaningful page numbers. This paper describes how the page numbers were dynamically assigned using SAS macro |
ods rtf, proc report, proc step, demographics table, mock shell, macro call, macro variable | source | 14 |
| 3 | A SAS Macro to Perform Bootstrapping for Internal Validation of Time-Dependent Area Under the Curve | Jialin Han, Jiali Ling, Maria E. Montez-Rath, Margaret R. Stedman |
We sought to develop an algorithm in SAS that uses the bootstrap for validation of a time- to-event prediction model and, in particular, to measure discrimination, which is the model's ability to distinguish between those who develop and those who do not develop the outcome of interest at a specific time t. The area under the curve at time t, AUC(t), is a statistic typically used to estimate discrimination, that can be implemented in SAS using the ROCOPTIONS command in PHREG. However, the AUC(t) is computed directly from the data used to develop the model so that overfitting is common. Overfitting occurs when the model is excessively complex, e.g., by including too many predictors and as a consequence, the model performs poorly when applied to other data. To assess the amount of overfitting, one can perform internal validation, using the same dataset, or external validation, using a new dataset. Here, we present a SAS Macro to compute a validated AUC(t) using the bootstrap method | sample, bootstrap, data cleaning, roc curve, auc, proc step, kaplan-meier estimator, proc phreg, cumulative frequency, probability distribution, estimate, macro definition, survival function | source | 9 |
| 4 | A SAS Macro to Perform Kaplan-Meier Multiple Imputation for Survival Analyses with Competing Events | Jialin Han, Sai Liu, Margaret R. Stedman, Maria E. Montez-Rath |
Analysis of cumulative incidence functions requires special attention when there are competing events. A competing event is an event that either precludes or changes the probability of the event of interest from occurring. There are different approaches to handling competing events in survival analysis. One approach, developed by Fine and Gray (1999), for example, cannot be applied directly to stratified data or to data with time- varying covariates. For these situations, Ruan and Gray (2008) proposed an alternative approach (Kaplan-Meier multiple imputation (KMI)) which recovers the missing censoring times for those who experienced a competing event. The missing censoring times are imputed from a non-parametric multiple-imputation approach based on the Kaplan-Meier estimator. In this paper, we introduce a user-friendly SAS Macro (%SASKMI) to implement this approach | imputation, proc phreg, proc step, proc calis, survival analysis, multiple imputation, competing risks, out=, macro definition, regression analysis | source | 8 |
| 5 | Alternate solution to %WINDOW statement in SAS Enterprise Guide: PowerShell | Sumit Pratap Pradhan | %WINDOW statement provides facility to define customized window that can be used to collect user input and display customized message. It is controlled by macro processor. SAS Enterprise Guide is a powerful windows application making SAS programming much easier and organized. Since SAS Enterprise Guide works on client/server model, it doesn't support %WINDOW statement. PowerShell is a powerful automation tool, consisting of command-line shell and scripting language. Windows PowerShell supports GUI programming. On a few occasions, SAS program needs to perform task dynamically based on the inputs provided by user. Since SAS Enterprise Guide doesn't support %WINDOW statement, an alternate solution was needed to encounter this scenario. I came across two solution s: 1-Use Prompt Manager to create a prompt which can be used to take user input; however, the drawback is that the Prompt Manager is available only if you create a project. So it doesn't suit to the requirement. 2-Use PowerShell | sas enterprise guide, txt | source | 5 |
| 6 | An Integrated-Software Solution for Automated and Reliable Distributed Reporting System | Noga Meiry Lewin, Aimee Wahle, Amarnath Vijayarangan, Abigail G. Matthews |
Ten years ago, o ur reporting team developed a web report system to present the status of ongoing clinical trials for the National Drug Abuse Treatment Clinical Trials Network. Different programmers contributed to a growing number of electronic data capture (EDC) platforms and studies, each with its own requirements, with all reports uploading to a common website. A common setup program standardized the programming process. Nightly one program ran to generate all reports for all platforms and studies, followed by a summary program that uploaded reports to the website. Any errors were investigated and corrected, and the program reran. As the number of active studies increased , the advantages of using one program to run everything were outweighed by generation time and debugging, and an error in one study prevented other studies from being updated | excel, edc | source | 11 |
| 7 | Analyzing Data Missing Due to COVID-19: From Raw Data to ADaM and Beyond | Yuliia Bahatska | The COVID-19 pandemic has a tremendous impact on everyone - and on the pharmaceutical industry. Naturally, those who are developing drugs took the first hit trying to treat it; however, statistical programmers who are working on any ongoing study had to reconsider their activities and timelines as well. New analyses need to be performed and the existing ones need to be amended taking lots of missing and incomplete data into account. CDISC has published the "Interim User Guide for COVID-19" [1], which describes how to represent COVID-19 related data with CDASH and SDTM. In my paper, I would like to go one step further and walk you through an example of preparing this data for analysis from mapping it to the VE SDTM [2] domain introduced by CDISC, to creation of the corresponding ADaM dataset, and then to the R shiny app supporting the analyses team was required to perform | create table, proc sql, adam dataset, clinical trial, cdisc | source presentation |
12 |
| 8 | Are You Qualified? Examining Existing SOC Occupational Groups and Assessing Candidate Suitability | Frank Yeboah, Maruthi Nanduri, Nikhil Gunti, Monish Vallamkonda |
Employee hiring process can be a cumbersome and tedious process with increasing number of job seekers entering the market. Adding to this a lack of transparency in the hiring process can put the candidates in the dark by not letting them know their shortcomings. In this paper, existing Standard Occupation Classification (SOC) groupings from the Bureau of Labor Statistics (BLS) are examined through hierarchical clustering based on descriptors that encompass individual skills and abilities. Later using O*NET occupational data, a hybrid recommendation engine on the LCA disclosure data of 2020 for H1-B and other programs is used to recommend the jobs for a candidate. The results of this paper reveal a gap within the current SOC groups and a need to re-evaluate the occupational groups based on the importance of skills, abilities, and other descriptors rather than only the description of the occupation | recommendation engine, hierarchical clustering, sample | source | 7 |
| 9 | Auditing Algorithms: The Need for New Tools to Combat Bias in Artificial Intelligence | Jason Brinkley | The field of algorithm development for machine learning/artificial intellig ence has seen rapid changes in recent years. And the rise in a lgorithm development has increased concerns of rampant bias in implementation. Many businesses turned to AI as a mechanism to create or evaluate things faster and cheaper with more accuracy than human beings. This hyper-focus on speed and accuracy has created numerous instances of AI with biases; there are many examples in the news and research articles of algorithms that have been shown to have strong gender or racial biases. There is a tug of war going on between the data science and policymaker communities around the need to better evaluate and audit AI to discover such biases before widespread rollout. The main issue is the need for methodological innovations and our existing tools are not well-suited for evaluating AI | machine learning, artificial intelligence | source presentation |
6 |
| 10 | Automated Vehicle Odometer Reading Using SAS for AI and Computer Vision | Tyler Rosacker, Kedar Prabhudesai, Hardi Desai, Neela Vengateshwaran, Varunraj Valsaraj |
Artificial intelligence (AI) is gaining increasing popularity with its ability to automate mundane, repetitive tasks that would otherwise require hours of manual work. In this paper we discuss one such application using SAS AI and computer vision to automa tically extract vehicle odometer readings from dashboard images. Reading vehicle odometers can be useful for verifying how far the vehicle has driven, and better understanding the risk the vehicle entails from an insurability perspective. However, with pot entially thousands of images received daily, it would not be practical for a human to conduct a review of each image. To overcome this, we have built a system using the You Only Look Once (YOLO) deep learning architecture, to automatically extract the odometer reading from a vehicle dashboard image | deep learning, insurer, confidence level, sample, histogram, type i error, auto insurance, model training | source |
17 |
| 11 | Back to Basics: Running an Analysis from Data to Refinement in SAS | Deanna Naomi Schreiber-Gregory | Data Science is the new Space Race, launching us into a world of immeasurable possibility, but with only a few people to help us navigate it. As we dig deeper, discover more, and risk more, we can be simultaneously led to both great insight and loss. If we do not know what we are doing, Data Science can be a very dangerous thing. It is important for us all to learn at least a little bit about the possibilities and risks of this field of study, so we can navigate it together. This paper was written to give individuals new to SAS and/or Analytics a gentle nudge in the direction of the possibilities available through Data Science and SAS. It is designed to help you navigate through the process of data exploration by using publicly available COVID 19 data. We have all seen how fragile this data reporting can be, and this paper uses this fragility to help explain the dangers of an inappropriately implemented analytic process | data science, nominal variable, missing value, null hypothesis, response variable, predictor variable, numeric variable, sample, sample size, logistic distribution, randomization, regression analysis, binary variable, interval variable | source presentation |
20 |
| 12 | Base SAS and SAS Enterprise Guide: Automate Your SAS World With Dynamic Code | Kent Ronda | Communication is the foundation of all relationships, including our SAS relationship with the Server, PC, or Mainframe. To communicate more efficiently and to increasingly automate your SAS World you will want to learn how to transform static code into dynamic code that automatically re-creates the static code, and then executes the re -created static code automatically. Our presentation highlights the powerful partnership that occurs when dynamic code is creatively combined with a dynamic FILENAME statement, the SET INDSNAME option, a Macro variable, and the CALL EXECUTE command within one SAS Enterprise Guide program node. You have the exciting opportunity to learn how to design dynamic code forward and backward to re-create static code while automatically changing the year |
filename statement, call execute, macro variable, sas enterprise guide, sas base | source | 23 |
| 13 | Bayesian Analysis of GLMMs Using PROC BGLIMM | Walter W. Stroup | Over the past two decades, generalized linear mixed models (GLMMs) have become standard tools for statistical analysis. Since its introduction, PROC GLIMMIX has been SAS/STAT's primary GLMM procedure. The most recent edition of SAS for Mixed Models includes three chapters on using GLIMMIX for GLMMs. PROC GLIMMIX is an excellent frequentist tool. However, Bayesian approaches are becoming increasingly important. Many academic journals prefer - some even require - Bayesian analysis. Even when not required, Bayesian methods allow you to use what you know prior to, or in the early stages of, an investigation. PROC BGLIMM is a new SAS/STAT procedure that makes Bayesian implementation of GLMMs relatively easy. BGLIMM uses syntax similar to PROC GLIMMIX, but there are some differences. This tutorial presents what you need to know to get started using PROC BGLIMM. We use GLMM examples from SAS for Mixed Models, but with a Bayesian twist |
proc step, proc glimmix, posterior probability, binomial distribution, sample, proc mixed, covariance, bayesian statistics, fixed effects, prior probability, response variable, repeated measures, likelihood function, least squares mean, regression analysis, gamma distribution, sample size, ods output, poisson distribution, proc reg, output, standard deviation, acf plot, class variable, credible interval, estimate, hypothesis testing, main effect | source |
25 |
| 14 | Best Practices When Moving SAS to Microsoft Azure | Margaret A. Crevar | The best practices presented here are lessons learned from SAS customers who have moved to Microsoft Azure. These best practices are being placed in a single location to share with other SAS customers looking to move their SAS applications to Microsoft Azure | source presentation |
5 | |
| 15 | Big Public Data | Laurie Smith | Big Data is everywhere and available to the public. It is eas y for us to access this data by use of Application Programming Interface (API). From United States census data to FacebookTM and Google TM data, t here is also much public data useful for studying Health Sciences. In addition to social media and other sites, healthdata.gov allows access via API to city, state, and country wide CoVID -19, Hospital- Acquired Infections, and Influenza Vaccination Rates data, amongst many other datasets. There are many health related datasets available to access through API. Analysis on this ty pe of data can be incorporated in many studies and can be accessed very easily using simple base SAS code | data dictionary, sas base, proc step | source |
8 |
| 16 | Bridging the Gap Between Legacy SAS and SAS Viya at the Centers for Medicare and Medicaid (CMS) | Richard Andrews, Manuel Figallo-Monge, Kevin Boone, Prakash Subramanian, Logan Perry |
SAS Viya has introduced many powerful capabilities to enrich the end user experience. From scalable infrastructure resources to easy-to-use front-end features, SAS Viya can improve a team or organization's ability to turn data into insights. Having decided to evaluate or move to SAS Viya, end-users are often left wondering how they can "bridge the gap" between their legacy investment in SAS to the more Cloud-native SAS Viya version. Based on the experiences of such users at CMS, this paper helps others facilitate a move to SAS Viya. Of particular interest are those end-users who: * Currently use client-server or mainframe applications --such as SAS Enterprise Guide, SAS Stored Processes, and SAS for the IBM mainframe-- for querying or coding and want to bridge the gap with their SAS Viya equivalents, i.e., SAS Studio and SAS Job Execution | sas viya, sas studio, machine learning, sas enterprise guide, cas, data governance, libname statement, parallel processing, sas programming, data science, in-memory analytics, sas model manager | source | 21 |
| 17 | Business Ethics at the Core of Successful and Sustainable Analytics | Mike McGuirk | The International Institute for Analytics has included analytics ethics as a top-five business imperative in both its 2019 and 2020 Analytics Predictions and Priorities list. Interestingly, this heightened need to focus on ethical practices coincides with a separate, related business priority to ramp up efforts to identify new ways to collect consumer data and apply analytical techniques on this data to gain a competitive advantage. In fact, in a recent study completed by Forrester Research, more than 50% of companies surveyed said they plan to launch initiatives to expand their ability to source external data. And many of these firms will appoint data hunters to lead these initiatives. In this session, we will dive into this critical moment in time for the analytics and data science communities. This is a time when an increased reliance on these practices will intensify the need to incorporate sound judgment and ethical business practices across all facets of analytics operations | credit card, customer experience | source | 2 |
| 18 | Copying Data Between SAS and JSON Files | Bruce F. Gilsen | JavaScript Object Notation (JSON) is an open standard file format and data interchange format used for some of the same purposes as XML. More information about JSON is readily available on the internet. Starting in SAS 9.4, you can copy SAS data sets to JSON files with PROC JSON. Starting in SAS 9.4TS1M4, you can copy JSON files to SAS data sets with the JSON engine. This paper provides basic information and some relatively simple examples. It also discusses ongoing research on how to copy JSON files into SAS in an automated way |
dataset, proc step, proc json, macro variable, sas library, character variable, libname statement, proc sql, json, length statement, numeric variable, proc datasets, ordinal variable, sysdate9, date variable | source presentation |
30 |
| 19 | Data-driven Programming Techniques Using SAS | Kirk Paul Lafler | Data-driven programming, or data oriented programming (DOP), is a specific programming paradigm where the data, and/or data structures, control the flow of a program and not the program logic. Often, data-driven programming approaches are applied in organizations with structured and unstructured data for filtering, aggregating, transforming and calling other programs. This paper and presentation explores several data-driven programming techniques that are available to SAS users | dictionary tables, proc step, proc sql, excel, proc contents, call execute, ods excel, sas library, sas log, output delivery system, macro facility, macro variable, dataset, select, character variable, proc format, proc print, options statement, user-defined format, proc freq | source presentation |
23 |
| 20 | Deep Learning to Classify Adverse Events from Patient Narratives | Qais Hatim, Tom Sabo |
Presentation: Can deep learning improve the efficiency of identifying safety signals from patient narratives? In this study of 14,976 narrative observations in clinical study reports where 50% of such reports include indications of serotonin syndrome, the most successful deep learning method tested achieved a 98.88% correct classification rate of serotonin syndrome. Furthermore, machine learning methods which provide a layer of interpretability including associated symptoms achieved a 94.4% correct classification rate. Deep learning and machine learning can improve the speed, accuracy, and interpretability of medical coding for adverse events. | adverse events, deep learning, machine learning, csr, clinical trial, data science, medical coding | source | 19 |
| 21 | Detecting Online Recruitment Fraud | Matthew Fruge, Noah Heinrich |
With the advent of the global pandemic caused by the Coronavirus disease, the job market is in flux, with many people currently unemployed. These individuals are seeking employment via job posting websites such as Indeed. Online job posting websites are vulnerable to fraudulent job posts created by scammers in order to socially engineer job seekers out of private or sensitive information. This paper examines job postings and attempts to automatically determine if the job posting is genuine (non-fraudulent) or fraudulent. Predictive models were used on job postings in the United States, labeling each as fraudulent or non-fraudulent based on variables provided in the dataset. By using a text topic node combined with a decision tree model in SAS Enterprise Miner, the team achieved over a 96% accuracy rate at distinguishing whether a job posting is fraudulent or legitimate | proc hpsplit, character variable, feature importance, data mining, binary variable | source | 6 |
| 22 | Developing a SAS Deep Learning MedDRA encoder (MedDRA-DeepCoder) for Patient Narratives | Qais Hatim | Patient narratives reported in clinical study reports (CSRs) provide clinical evidence of adverse events that occurred to a patient and help scientific reviewers during pharmacovigilance activities. The manual review of these narratives is a daunting task for safety reviewers as it is time consuming and resource intensive. How can we improve the efficiency of identifying safety signals from patient narratives? Can deep learning technology help to overcome the review challenges in an automated way? This paper suggests an implementation to accurately categorize one adverse event term, "Serotonin Syndrome", as an example of what SAS deep learning technology is capable of. We first generate sentence level embeddings from terms contained in patient narratives. Following this, we generate term embeddings within a SAS deep learning framework. Subsequently, we obtain a category decision on whether or not the narrative text relates to Serotonin syndrome as an output | deep learning, adverse events, neural network, machine learning, data cleaning, validation set | source | 11 |
| 23 | Disproportional Impact of COVID-19 on Marginalized Communities | David J. Corliss | The impact of COVID-19 pandemic has been especially severe among marginalized communities, including persons of color, Native Americans, LGBTQ+, the homeless, and prison populations. This study leverages county-level COVID mortality data from the New York Times database and socio-economic and demographic data from the American Community Survey to identify risk factors and calculate odds ratios for COVID-19 mortality among various marginalized communities. Disproportionate impact is assessed using only the first wave of the pandemic in order to measure intrinsic risk unaffected by politically- driven behaviors and associated mortality developing in later waves of the pandemic. Per capita mortality is calculated for sets of counties with 60%+ population of at-risk groups and compared using odds ratios to the set of counties with less than 5%. These risks are compared to other COVID-19 mortality risk factors, e.g., chronic lung disease | odds ratio, time series analysis, proc step | source presentation |
10 |
| 24 | Do Americans Trust Scientific Experts? | Jaswinder Singh, Kalyan Alisetty, Vinil Rayala, Taylor Rebollar |
This report presents the results from an analysis of Americans' Mistrust in Scientific Experts. PEW Research conducted a survey asking the general American public questions about their views and opinions regarding their level of trust in scientific experts. The data includes variables such as demographics, responses to survey questions intended to measure the level of trust, and responses to factual questions to assess knowledge. The various categories of scientific experts used in this study were the following: Medical Doctor; Medical Res earch Scientist; Dietician; Nutrition Research Scientist; Environmental Health Specialist and Environmental Research Scientist. This report presents results from descriptive analyses that highlight trends based on the responses given. Predictive models are also presented which can take responses from an individual and predict their level of trust in a scientific expert of a certain category | response variable, missing value | source | 7 |
| 25 | Do I Have Enough? - A Macro for Sample Size Determination in Simple and Multiple Logistic Regression | Carl P. Wilson | Binary logistic regression finds plenteous usage throughout many scientific disciplines. Despite its multitudinous applications, there lacks a universal method of determining the sample size for a binary logistic regression model. Some suggest using 'magic number' methods to estimate the sample size, but such methods do not include statistical power, effect size, or significance level into their approximations. To remedy this dilemma, this paper presents a SAS macro that automates the sample size determinations for both simple and multiple binary logistic regression models with continuous predictors. This macro- based on Hsieh's (1998) formula-lacks the limitations from which the magic number method tolerates. Researchers will benefit from using this macro to determine the true sample sizes for their studies and to overcome the imbroglios of under and overestimating a sample size | sample, sample size, logistic regression, regression analysis, logistic distribution, effect size, significance level, standard deviation, response variable | source sas |
10 |
| 26 | Embracing Digital Transformation | Ben Zenick, Christopher Beall |
Zencos and CWE work together to build solutions for organizations that leverage SAS software and cloud-native capabilities for the tasks of data retrieval, consolidation, and visualization. Today's paper will walk through how digital transformation and cloud analytics can help organizations achieve quicker time to value by providing insights into technical operations that expose measurable cost reduction opportunities. | data cleaning, least privilege | source | 9 |
| 27 | Fundamentals of Machine Learning | Labeeb Khan | Presentation | machine learning, unsupervised learning, data cleaning, model drift, data drift, discriminant analysis, type ii error | source | 27 |
| 28 | Generating .xpt files with SAS, R and Python | Todd Case | The primary purpose of this paper is to first lay out a process of generating a simplified Transport (.xpt) file with RStudio and Python to meet study electronic data submission requirements of the Food & Drug Administration (FDA). The second purpose of this paper is to compare the .xpt files created from three different languages: R, Python and SAS. The paper is the expansion of the original FDA guideline document "CREATING SIMPLIFIED TS.XPT FILES", published in November, 2019. Transport files can be created by SAS, as well as open source software, including R and Python. According to the FDA guideline document mentioned above, .xpt files can be created by R and Python. This may allow Pharmaceutical companies to expand use of R and Python beyond data visualization and statistical analysis currently being generated by these two languages. Hopefully, readers can use the process shown in the paper as a template to create .xpt files | dataset, trial summary, data science | source | 11 |
| 29 | Getting Started with the R SWAT Package in SAS Viya | Brian K. Varney | SAS Viya gives SAS and R developers the ability to collaborate and work off the same data sitting in memory on a SAS Viya server. This paper intends to show how to set up the connection from RStudio and process data using SAS Cloud Analytic Services (CAS). Examples will also be shown comparing the execution of analytics using R Studio and SAS Viya | sas viya, cas | source | 9 |
| 30 | Git and SAS: A Match Made in (SAS) Studio | Joe Matise | Once upon a time, version control was something for programmers who lived on the command line, and required remembering esoteric commands and complex sequences. Thankfully, this is no longer the case - with the latest version of SAS Studio, version control using Git is now easily accessible for all! We will cover the basics of using Git integration in SAS Studio, from how to set up a repository, how to connect it to your company's server, how to commit changes to your code, and how to view past versions of your code. For those of you who are seasoned Git pros, we will show how Git integration in SAS Studio can save you time, making for a more concise and smooth workflow while maintaining your ability to use Git outside of SAS Studio for those more advanced operations you need it for. This presentation is aimed at programmers at any level who either know nothing at all about Git, or who would like to learn how to use the built-in Git functionality in SAS Studio | sas studio, sas viya, sas programming | source | 11 |
| 31 | Harnessing Raw Data From Submission Forms to Gauge Food Insecurity Among College Students | Zoraya Cruz-Bonilla | Submission forms are a common practice for collecting student information quickly and directly. The pandemic outbreak has spurred this practice even further in order to pivot in - person campus operations to a safer environment or platform. However, not all submission forms are created equal. Some lack adequate validation rules to impose data integrity; while others fall short on their promise to provide greater meaning due to their flat architecture. Thus, valuable insights easily get lost among the hundreds of records generated. Through the integration of SAS Enterprise Guide and SAS Visual Analytics, participants will learn a handful of functions to consolidate raw data from a series of weekly CSV files; they will learn how to create a handful of custom fields to underscore service usage trends; and how to create on-demand reporting. The outcome is a transformative process that renders raw data to information that is actionable and supports public service efforts |
raw data, sas enterprise guide, by group processing, binary variable, data cleaning | source presentation |
12 |
| 32 | Haste Makes Waste: Don't Ruin Your Reputation With Hasty Regression | Steven C. Myers | How do you know if your data is lying to you? The answer lies in following ethically applied econometric rules and being aware of and avoiding pitfalls in regression practice. These essential skills are not typically taught in a single-semester econometric or regression course. This paper defines hasty regression and demonstrates why it represents a terrible testing strategy leading to unsubstantiated and false conclusions. In this paper, a baseline regression includes a dummy variable for a treatment. The hypothesis is that this policy variable represents a statistically significant change in the outcome measure of the dependent variable. A simple data set is used to illustrate that the initial insignificant result is dead wrong and in the end, has an outsized and positive effect that is superior to alternative specifications. None of this would be known if the analyst stopped after the false signal of that first hasty regression | regression analysis, null hypothesis, sample, proc step, proc reg, response variable, binary variable, machine learning, proc autoreg, feature selection, data science, nominal variable | source presentation sas |
39 |
| 33 | Hello Graphics World: Basic Charts in SAS, R and Python | Ted Conway | In this presentation, we'll take a look at what it takes to generate fundamental chart types using SAS (ODS Graphics), R (ggplot2) and Python (Matplotlib). Attend this session to get a feel for some of the pluses and minuses of each language's approach. You'll see that there's often no need to leave the friendly confines of Base SAS to address your charting needs! |
histogram, data visualization, proc sgplot, bubble plot, line chart, scatter plot, proc step, box plot | source presentation |
11 |
| 34 | How Many Shades of Guide: SAS Enterprise Guide to 8.3 and SAS Studio to 3.81 with SAS 9.4 | Philip R. Holland | I have been using SAS Enterprise Guide since version 1.1.1 in 2001, and SAS Studio since version 3.1 in 2013. SAS Enterprise Guide version 8.3 and SAS Studio version 3.81 have developed immensely from their original releases. This paper tracks their progress and convergence towards the perfect SAS development environment. If you are not yet using SAS Enterprise Guide or SAS Studio, then your IT department may be unnecessarily overworked, and you are missing out on some very useful programming features | sas enterprise guide, sas studio, dataset, sas programmer, sas log | source presentation |
11 |
| 35 | How SAS 9.x enriched your SAS programmer life? | Angelo Tinazzi | SAS9 is the longest-running version as it was first released in 2002, with its last technical maintenance release last summer (SAS 9.4 M7). In this presentation, I will go through some of the main Base SAS language enrichments you might have missed. For example, new functions to handle character variables, new DATA STEP options and new macro features. The focus will be on improvements that make work life better for me and every other SAS programmer |
sas 9, hash table, proc step, macro variable, proc sql, dosubl, macro definition, call execute, macro facility, character variable, character functions, %if-%then-%else, if-then-else | source |
15 |
| 36 | How Sick Is My Cohort Of Patients? A General Approach to Identify Chronic Conditions | Patricia Ferido, Laura Gascue, Patricia St. Clair |
Analysis of medical claims data frequently requires identifying a disease cohort or controlling for chronic conditions. The Chronic Conditions Warehouse (ccwdata.org) applies validated algorithms for discerning multiple chronic conditions in Medicare claims data, and CMS provides the results of their application with Medicare claims data if requested. The algorithms use a combination of diagnosis and procedure codes and selection rules, which vary among conditions. For example, to eliminate rule-out diagnoses, the algorithm for diabetes requires one set of diagnosis codes observed on any inpatient or skilled nursing facility claim, or two claims at least a day apart on a physician claim. This presentation discusses a SAS macro that applies CCW-like rules to any data set either from insurance claims, or electronic health records (EHR) containing a full picture of diagnoses and procedures from patient medical visits | insurance claims, sample, data cleaning, macro function, dataset, excel | source | 12 |
| 37 | How to Deal With Locked Data Sets on UNIX systems | Kurt Bremser | This paper presents a method to help in updating datasets that are in use by another process | macro loop, macro variable, %if-%then-%else, macro definition | source presentation |
5 |
| 38 | If You Need These OBS and These VARS, Then Drop IF and Keep WHERE | Jayanth Iyengar | Reading data effectively in the DATA STEP requires knowing the implications of various methods, and DATA STEP mechanics; The Observation Loop, and the PDV. The impact is especially pronounced when working with large data sets. Individual techniques for subsetting data have varying levels of efficiency and implications for input/output time. Use of the WHERE statement/option to subset observations consumes less resources than the subsetting IF statement. Also, use of DROP and KEEP to select variables to include/exclude can be efficient depending on how they're used |
data=, dataset, drop statement, if-then-else, where statement, subsetting if, cpu time, set statement | source | 9 |
| 39 | Improving Root Cause Analysis and Mean Time To Resolution in complex SASEnvironments | Christopher Blake | Root Cause Analysis (RCA) is a common IT practice aimed at identifying the underlying causes of faults and problems with systems with the objective of resolving them quickly to ensure system uptime and customer satisfaction. When the results of this are measured, this is often referred to as Mean Time To Resolution (MTTR). While common in IT generally, Root Cause Analysis is often not performed on SAS platform issues which can have a direct impact on system stability and overall user satisfaction of the system. The variety of workload types, number of servers involved, and the amount of moving parts can make performing Root Cause Analyses of SAS platforms very difficult | root cause analysis, customer satisfaction | source presentation |
9 |
| 40 | Introduction to Azure Kubernetes Service and SAS Viya 2020 | Vasilij Nevlev | With the increasing acceptance of the cloud, there has been a massive shift in how software is designed, built, and deployed. Developers have turned to containerisation to make it easier and faster to create new products and release them. This paper has been written for people unfamiliar with terms such as Kubernetes and Docker containers. It discusses how SAS Viya 2020.1 uses these through Azure Kubernetes Service. By the end of this paper, you will be equipped with enough basic knowledge to understand technical deployment documentations. | sas viya, sas enterprise guide | source |
6 |
| 41 | Introduction to Doxygen | Tom Bellmer | A good programming practice is to document your code via the use of comments in the header area. Comments support the understanding of code when you have been away from it or when others need to support it. Generating external documentation from existing code comments is highly desirable as to do otherwise will inevitably cause things to no longer be synchronized. A free open source utility named doxygen does what is needed and this paper describes how to use it | ods html, proc sgplot | source | 11 |
| 42 | Investing in Education Intelligently | Sam Edison, Rushya Puttam, Akhil Emani, Nithisha Katta |
Student loans are a critical resource used by many Americans to help pay for the goal of achieving higher education. As the tuition for post-secondary education continues to rise, the need for understanding student loans and the variables that impact individuals' abilities to pay them back is vital. In a recent survey, Country Financial found that only 9% of parents talk to their children about managing student loan debt (PR Newswire, 2019). Many borrowers are teenagers who have little formal financial education and even less experience in such financial matters. The College Scorecard is an online tool assisting students to evaluate higher education options. Created by the United States government, the tool helps address the disconnect experienced by future students and the high cost of advanced education. Using SAS, this study analyzes data compiled by the Scorecard to provide further insight into possible student outcomes based on institution-level data | response variable, feature selection, regression analysis, missing value, neural network, proc hpforest, correlation, nominal variable, interval variable | source presentation |
7 |
| 43 | It's in My DNA: A SAS Program to Assist With Finding Unknown Relatives | Richard D. Langston | This paper describes a SAS program I created to process DNA matches from Ancestry.com and which assists the user in creating descendancy listings to help determine previously unknown biological parents. The program reads HTML data to extract match results, and provides a mechanism to cross-reference these results with relationships constructed from family trees, obituaries, and other historical records | ods html, numeric variable | source | 12 |
| 44 | Large-scale demand forecasting for e-commerce retailer using SAS Forecast Studio: A Case Study | Ajay Kumar Mishra, Mohak Narkhede, Padmanabha Chandu Saladi |
A leading e-commerce retailer with multiple websites desired to create a systematic, scalable and science-driven product forecasting system. We configured SAS Forecast Studio for this client on a cloud-based server for automated weekly forecasting of tens of thousands of products for the next year. We used SAS for data insights like understanding the distribution of product life and rate of sale, variability in demand and prices, as well as seasonality among the products. Pre-processing included creating demand history by adjusting order history for lost sales using inventory data. Price is an important predictor of demand for this retailer. We modeled holiday events using HPFEvents in SAS Forecast Server Procedures. With a deep understanding of the patterns in the data for this retailer, we configured SAS Forecast Studio including use of test runs to select independent variable form, fit criteria, reconciliation, and hierarchy choices | predictor variable, time series analysis, proc esm, residual, sample, data science | source | 9 |
| 45 | Let's enjoy the computer vision using SAS Viya! | Shinjo Yada, Ryuji Uozumi |
Image recognition technology enables machines to understand the content reflected in an image. Many studies in recent years have applied machine learning methods to computer vision problems. Among these methods, convolutional neural networks (CNN) is the most commonly used. CNN also has applications in other fields, such as natural language processing. Using CNN, higher-order features can be extracted by stacking layers that perform two different modes of processing. This report provides an illustration of the CNN structure. Further, it will address regions with CNN (R-CNN), used for object detection, and its successors, Fast R-CNN and Faster R-CNN, as examples of the application of CNN, as well as provide a schematic explanation of their processing flows. Lastly, this paper will introduce the program code of Faster R-CNN using SAS Viya | neural network, regression analysis, sas viya, deep learning, machine learning, posterior probability, network optimization | source | 19 |
| 46 | Log Reviewing Made Easy | Peter Knapp | One of the golden rules when running SAS programs is to review the Log for problems. Unfortunately, many users ignore the Log and focus on the output. Adding a dynamically generated report to the bottom of the Log that shows the number of problems (such as run- time errors, missing values, and uninitialized variables) is a great way to identify issues with the program run that need addressing. Enhancing the report to include information about the number of observations flowing through the program can not only help ensure that the program runs without mechanical errors, but also runs as expected. This presentation discusses a Log Report macro that provides summary log information, explains how it works, and details how to modify it so you can tailor it for your organization's specific needs. The Log Report was developed using SAS Enterprise Guide on a Windows environment using SAS macro and Base SAS code. The Log Report also works in SAS Studio | put statement, sas enterprise guide, macro variable, sas studio, macro put, error message, call symputx, sample, macro definition | source presentation |
10 |
| 47 | Making SAS Tables Smaller and Faster Without Data Loss | Glen Becker | This session explores how SAS physically stores and retrieves data from SAS data files. It describes SAS data compression and other space-saving techniques. It explores the performance storage implications of SAS data and SQL views, and demonstrates how views can save space and time. It also explores the performance implications of compression and the choice of variable length, including a surprising technique that can make a SAS table smaller by making some variables larger. Lastly, it introduces "Zipper", a SAS macro that zips SAS data, and builds a SAS data view that expands the zipped data on-the-fly | character variable, dataset, proc step, missing value, numeric variable, proc sql | source presentation |
16 |
| 48 | Making Text Analytics More Approachable | Sundaresh Sankaran | Text Analytics proves most effective when it is approachable to business users, packaged within easily deployable and accessible applications, and helps you make better and effective decisions at scale . Core text analytics capabilities, including those offered by SAS , help analysts and data scientists categorize documents, extract information, discover topics and identify sentiment. In this paper, we offer a selection of applications built over these core capabilities, each of which help you obtain answers to questions that motiva te you as a business users of AI , and enable you to receive recommendations, gain insight, and make decisions. Our hope is that these applications serve to expand your understanding of the applicability of text analytics capabilities, and the value it brings to your organization | data science, sas viya | source | 9 |
| 49 | Map It Out: Using SG Attribute Maps for Precise Control of PROC SGPLOT Output | Joshua M. Horstman | The SGPLOT procedure, part of the ODS Statistical Graphics package, allows for extensive customization of nearly all aspects of plot output. These capabilities are commonly used to distinguish between groups or categories being compared through the use of distinct plot attributes, such as symbols and colors. However, there are times when it is advantageous to be able to associate specific plot attributes with specific data values. SG attribute maps provide functionality that does exactly that. This presentation will provide an introduction to the use of SG attribute maps in conjunction with PROC SGPLOT. A series of examples will demonstrate how attribute maps are used and why they are useful as a programming tool. Both discrete and range attribute maps will be used to modify a variety of plot attributes, such as plot marker symbols and colors, line styles and fill patterns |
scatter plot, proc sgplot, proc step | source | 11 |
| 50 | Methods of a Fully Automated CONSORT Diagram Macro %CONSORT | Jeffrey P. Meyers | The CONSORT diagram is commonly used in clinical trials to visually display the patient flow through the different phases of the trial and to describe the reasons patients dropped out of the protocol schedule. They are very tedious to make and update as they typically require using different software, such as Microsoft Visio or Microsoft PowerPoint, to manually create , align, and enter the values for each textbox. There have been several previous papers that explained methods of creating a CONSORT diagram through SAS 1, but these methods still required a great deal of manual adjustments to align all of the components. The %CONSORT macro removes these manual adjustments and creates a fully automated yet flexible CONSORT diagram completely from data. This presentation is a description of the | group by, adverse events, data step, proc sql, macro call, macro variable, character variable, unique key, proc freq, macro loop | source sas txt presentation |
20 |
| 51 | Midsize Banks and Credit Unions Level the Playing Field With AI and Advanced Analytics | Larry Hall | Presentation | source | 31 | |
| 52 | Midsize Banks and Credit Unions Level the Playing Field With AI and Advanced Analytics | Lawrence Dennison-Hall | Presentation: The main challenge for many banks today is finding ways to remain competitive in a rapidly changing market. A market that includes increasing competition from both new and traditional competitors, a demand for hyper personalized services from clients, and difficulty in distinguishing products and services from the competition. Larger banks are busy building out their own data science teams and even buying entire companies that bring advanced analytics and AI to their internal operations and client offerings. For midsize banks without unlimited resources there is a solution: SAS software. Using SAS, these banks can cost-effectively use analytics and AI to not just level the playing field, but also leapfrog big banks in areas like operational cost reduction, customer experience, and client acquisition and retention | machine learning, financial services, artificial intelligence, customer experience | source | 9 |
| 53 | Modeling Causal Effects Using SAS/STAT | John Amrhein, McDougall Scientific |
In a broad sense, correlation is any relationship between two variables. Causation is a stronger relationship because the value of one variable induces the second variable to take its value. Causation can be quantified when the two variables are measured close in time, but with the cause always preceding the outcome. To establish a causal relationship, we must isolate the cause and outcome from other possible causal factors by intervening in the world we are measuring. To say this another way, causation is all about understanding the process that generated our data. It is not enough to discover relationships in the data, we must understand how those relationships were created. This paper is a gentle introduction to causal inference using the method of structural causal models. The first section defines causality and describes the conditions necessary to declare causality from an analysis. This is followed by an introduction to directed acyclic graphs, or DAGs, which are instrumental to understanding and fitting structural models. Using the SASHELP.HEART data set, a DAG is constructed to support several analyses. For several causal factors of interest, we use the CAUSALGRAPH Procedure to identify adjustment sets containing factors that confound the causal factor of interest. Having identified the confounders, we use the CAUSALMED and CAUSALTRT Procedures to fit our structural causal models to estimate causal effects. |
proc step, treatment effect, potential outcomes, label statement, response variable, propensity score, randomization, regression analysis | source | 21 |
| 54 | Modernizing Scenario Analysis With SAS Viya and SAS Visual Analytics | Arne Cordes, Stu Sztukowski |
Data scientists and statisticians spend months perfecting their models and presenting results. Whether it's running in SAS, Python, or R, model results need to go somewhere where someone can consume them - even the best model has little value if a decision maker can't see them. Dashboards or daily emailed reports are usually the most natural place for model results to go. To make greater use of models, decision makers need more than just the daily numbers. They need to find out how the results will change under varying conditions. This task is trivial for a data scientist: simply write a program to change the numbers and run a scenario; however, not everyone has that skillset. To quickly allow others to create scenarios, data scientists typically would place parameter estimates within a carefully crafted spreadsheet and send it to decision-makers | scenario analysis, macro variable, sas studio, missing value, response variable | source presentation |
21 |
| 55 | Montreal Cognitive Assessment, Mini-Mental State Examination: Impairment & Dementia Cutoff Points | Feier Han, Zhen Wu, Jing Shang, Yutong Zhang |
As the population ages with increased life expectancy, the prevalence and incidence of dementia and mild cognitive impairment (MCI) continue to increase. About 5-8% of Americans aged 60 or above are diagnosed with dementia (WHO, 2020), and 6.7-12.5% of them have MCI (Petersen et al., 2018). The increasing prevalence of MCI and dementia has resulted in increasing comorbidities and the associated healthcare expenditure. Moreover, even with comprehensive treatments, MCI and dementia can significantly impact both patients' and caregivers' quality of life. As such, it is critical to diagnose cognitive impairment and prescribe interventions promptly. A good screening test can significantly improve the efficiency and accuracy of MCI and dementia diagnosis. The Mini-Mental State Examination (MMSE) and the Montreal Cognitive Assessment (MoCA) are two commonly used cognitive function screening tools specifically developed for screening MCI (Folstein et al., 1975; Nasreddine et al., 2005). Previous studies had investigated the optimal cutoff scores of MMSE and MoCA for MCI and mild Alzheimer's disease among elderly Chinese veterans (Tan et al., 2015) and in a Singapore population (Ng A et al., 2013) according to age and education level. Additionally, Chapman's work (2016) focused on a large Alzheimer's cohort to explore MMSE cutoff thresholds for clinical trials and diagnostic use, also stratified by age and education level. However, these studies were not conducted in the general population and some had a relatively small sample size. It is essential to have a clinically useful diagnostic test for cognitive status in various subgroups. | sample size, sample, roc curve, logistic distribution | source presentation |
8 |
| 56 | Moving SAS Viya Contents: A RESTful Way | Amit Choubey, Sounak Nag |
A SAS Customer wanted an application to be built around SAS Viya, which would help their IT organization to manage their Change Management process. A solution was needed to help their developers promote SAS Viya contents from one environment to another without logging into the SAS Viya user interface and without having to learn the SAS export/import tools. Due to their Group IT Policy, the customer was reluctant to use the sas-admin CLI and wanted to use the REST APIs to promote the contents from one environment to another. We explored the SAS Viya REST API's and built an application that calls and orchest rates the promotion process in the backend to export and import of SAS Viya contents from one SAS Viya environment to another. The application calls the SAS Viya promotion APIs to move the contents across the environment at the user -provided date and time to promote the contents to higher environment | rest api, sas viya | source |
9 |
| 57 | My First SAS Package: A How-To | Bartosz Jabłoński | A SAS◦Rpackage1 is an automatically generated, single, stand alone zip file containing organised and ordered code structures, created by the developer and extended with additional automatically gener- ated "driving" files (i.e. description, metadata, load, unload, and help files). The purpose of a package is to be a simple, and easy to access, code sharing medium, which allows: on the one hand, to separate the code complex dependencies created by the developer from the user experience with the final product and, on the other hand, reduce developers and users unnecessary frustration related to a remote deployment process. The SAS Packages Frameworkis a "pack" of macros, which allows touse and todevelop SAS packages | sas packages, sas base, proc step, macro definition, macro put | source presentation |
13 |
| 58 | Natural Language Processing: An Introduction | Colleen M. Farrelly | Natural Language Processing (or NLP) is a growing field of machine learning concerned with understanding text data and integrating insights from text data with other data sources. Often, insights contained in text data aren't captured in relational data or other sources. Through processing and analyzing text data, these insights can be integrating wit relational data within a single model, dashboard, or other analytics tool | structured data, relational database | source | 4 |
| 59 | Next-Generation Data Mining and Machine Learning: Python and SAS Together at Last | George S. Habek, Suman Kumar Das |
Have you developed Python models to solve a business problem and wish to leverage that asset into a next generation analytical framework? Learn how to create a bridge between open source, using Python, and a state-of-the-art data mining & machine learning environment, using SAS VDMML, which is automated, dynamic, and scalable - right at your fingertips. Compete and compare next generation algorithms to select a champion seamlessly. The business problem will be within the transportation shipping industry. A variety of data elements such as internal factors of package size, delivery day of the week, # transit days; external factors of weather and financial effects will also be used. The ultimate business question - What is the best subset of factors that drive cost? What is the optimal model to use to score a new prospect file with the predicted average cost? Discover how Python and SAS VDMML can work together to answer these critical questions for the shipping company! | proc hpforest, proc hpsplit, proc reg, neural network, regression analysis, sas visual data mining and machine learning, machine learning, champion model, network optimization, sample, data mining, feature importance, elastic net, data cleaning, supervised learning | source presentation |
24 |
| 60 | Open Source SAS Macros - What? Where? How? | Katja Glass | Open Source provides a lot of opportunities to work smarter. This presentation will give an overview about open-source solutions which are available for SAS. The presentation is going to briefly explain how open-source can be applied in companies and organizations, what licenses are available and what other important aspects to consider for example when it comes to maintenance, documentation and validation. The core of the presentation will be the overview of the various solutions to allow a brief overview of what is available. These are SAS macros, scripts and related tools | clinical trial, data cleaning, excel | source presentation |
15 |
| 61 | Pandemic Pandemonium | Hannah Flynt, Maryam Taherirani, Sean Everett, Trinh Phan |
Losing a job is not only devastating to the individual and affected household, but is also a major economic issue. Okun's law, named after Yale Economist Arthur Okun, states that even a 1 percent increase in US unemployment causes a reduction in US GDP by 2 percent (Kenton, 2020). With the recent global health crisis, job loss has drastically increased and the goal of this paper is to explore publicly available demographic and infe ction rate data, in order to understand the variables that most impact job loss and to predict the likelihood of losing one's job. Whether the household has experienced job loss is the target variable. The final model, built in SAS Enterprise Miner, was a logistic regression model with a misclassification rate of 35%. Some identified variables of importance were Education, Income, Age, Health Insurance Coverage, Metropolitan Statistical Area, Government Response Index, Infection Rate and Number of Adults in the Household | odds ratio, regression analysis, data cleaning, logistic regression, anova, data mining, logistic distribution | source | 7 |
| 62 | PKSI Advanced Banking Analytics | Larry Hall | Presentation | customer experience, machine learning | source | 6 |
| 63 | Prediction of Parkinson's Disease With a Regression Model Using Vertical Ground Reaction Force Data | Jin Ge, Tianchen Xu, Jia Guo, Mengyu Zhang |
Parkinson's disease (PD) is a long-term degenerative disorder of the central nervous system. Approximately 60,000 Americans are diagnosed with PD each year. The Parkinson's Foundation Prevalence Project estimated that 930,000 people in the United States were living with PD by the year 2020. This number is predicted to rise to 1.2 million by 2030. Patients with PD may experience movement or motor related difficulties such as tremor, slowness of movement (bradykinesia), rigidity, and postural instability. Our goal is to investigate the most influential features, including demographics and motor related biomedical signals, for PD prediction. PD gold standard for diagnosis is based on subjective clinical evaluation. An objective and quantitative gait analysis system could potentially improve the current practice in diagnosis, symptom monitoring, therapy management, rehabilitation and fall risk assessment and prevention in PD patients. | logistic distribution, time series analysis, logistic regression, standard deviation, proc step, regression analysis, proc princomp, confidence interval, roc curve, test set, proc freq | source presentation |
7 |
| 64 | PROC HTTP: Using the Web to Combat Fraud in Auto Finance During a Global Pandemic | Joshua Piers | Most car dealerships do not buy cars outright; they finance them until they sell to a consumer. While the vast majority of dealerships conduct themselves respectably, there are occasional cases of dealership fraud. Typically, auditors are sent to dealerships to monitor vehicle inventory and verify cars are on the lot. In the midst of a global pandemic, this has been difficult to do effectively. The practice of selling vehicles without repaying the lender allows a dealership to float cash since it typically only has to pay a monthly interest amount as opposed to the full amount for the vehicle when it sells. Compounded, this practice can set a dealer back beyond the point of return, potentially resulting in hundreds of millions of dollars in loss. Time to get creative! Using PROC HTTP, it's possible to grab HTML code from the dealer inventory web pages, identify VIN numbers from the HTML, and compare the list of VINs to the lender's active loans |
proc http, proc step | source presentation |
7 |
| 65 | Quick, Call the "FUZZ": Using Fuzzy Logic | Richann J. Watson, Louise S. Hadden |
SAS practitioners are frequently called upon to do a comparison of data between two different data sets and find that the values in synonymous fields do not line up exactly. A second quandary occurs when there is one data source to search for particular values, but those values are contained in character fields in which the values can be represented in myriad different ways. This paper discusses robust, if not warm and fuzzy, techniques for comparing data between, and selecting data in, SAS data sets in no t so ideal conditions | proc step, where statement, index, fuzz, decimal, if-then-else, prxmatch, proc fcmp, upcase function, sample, character variable | source | 30 |
| 66 | Reliability of Individual Predictions Estimation for Claim Risk-Scoring Models | Yuriy Chechulin, Jina Qu, Hudson Mahboubi, Terrance D'souza |
The Workplace Safety and Insurance Board (WSIB) is an independent agency that administers compensation and no-fault insurance for workplaces in Ontario, Canada. The WSIB develops claim risk scoring models using SAS to identify high-risk claims. Standard methods for assessing predictive model accuracy estimate an overall predictive power of the model (i.e. accuracy, C-statistic) averaged across all test observations. Even if the model is very accurate, on average, across all observations, it does not mean that individual predictions have the same reliability across the data. If an individual prediction for a person who has been injured at work has low reliability, the default action suggested by the model (i.e. triage into high versus low-risk group) could be ignored and additional human intervention can be taken | roc curve, survival analysis, scatter plot, proc hpforest, auc | source | 9 |
| 67 | Reporting Correlation Coefficient Results and Plots: A SAS Macro That Does It All | Zhengming Chen, Bowen Zheng |
In academic and particularly medical research, the essential first set of results are often summarized in a table called Table-1. Table-1 is an informative cross-table that consists of a set of numerical and categorical variables in rows stratified by a categorical variable in the columns. This single table provides the appropriate summary statistics for all numerical and categorical variables assessed in the study as well as their bivariate associations with the categorical variable that they are stratified by. There are many user-written SAS Macros dedicated to producing such a Table-1 with minimal effort. However, these cross tables cannot provide the bivariate associations between two numerical variables. To supplement these Table-1s, this paper provides a SAS Macro that, with minimal input, produces a publication- ready document that reports in a single table the summary statistics of numerical variables, an appropriate correlation coefficient between two numerical variables with its confidence interval, and the p-value from hypothesis testing for association. To assist in the interpretation of results, it can also optionally produce footnotes and publication-quality figures to visualize the relationship between variables. This SAS Macro is simple and easy to use and it provides a one-stop, powerful tool for common types of correlation studies between two numerical variables. It has been fully tested in different SAS interfaces and is incorporated with the SAS ODS system. |
correlation, macro definition, pearson correlation, null hypothesis, regression analysis, scatter plot, sample, proc step, hypothesis testing, ordinal variable, confidence interval, proc corr, standard deviation, proc means, interval variable, maximum likelihood estimation, output delivery system | source presentation |
11 |
| 68 | Running SAS Workloads Using Clustered File Systems on Amazon Web Services | Dilip Rajan | High-performing SAS applications in the cloud have different design patterns and storage options. Having the right architecture is critical in a customer's transition to the cloud, as setup and maintenance of poorly performing applications can be time-consuming and expensive to mitigate. With Amazon Web Services, customers have a variety of storage and compute options, but getting the right compute and storage configuration is vital to ensure success and a smooth transition. Amazon FSx and Amazon EFS provide simple, scalable, fully managed network file systems that are well-suited to provide the shared storage required for most customer applications. For customers that want to lift and shift their existing on-site SAN architecture to AWS without refactoring their cluster-aware file systems, such as Red Hat Global File System 2 (GFS2) or Oracle Cluster File System (OCFS2), another option is to use Amazon EBS volumes and the Multi-Attach feature | sas grid, amazon s3, red hat | source | 8 |
| 69 | Running SAS Workloads Using Clustered File Systems on Amazon Web Services | Dilip Rajan | Presentation | analytics platform | source | 13 |
| 70 | SAS Deployment on Google Cloud Tips/Tricks including BigQuery Integration | Ande Stelk | A technical discussion on how to best deploy SAS in GCP touching on Google Compute Engine, Storage and other configuration tips and tricks would be included. We will also cover integration with BigQuery using SAS Access to BigQuery. Viya v3 and v9 platform s would be referenced. This would be a specific presentation to GCP | sas viya, sas grid, cloud storage, proc sql, sql pass-through | source | 8 |
| 71 | SAS Time Series Analysis & Forecasting (TSAF) at the Canada Revenue Agency (CRA), with COVID impacts | Jason A. Oliver | It may well be a recurring theme of this year's SAS Global Forum that we are faced with more pressure to use flexible thinking - not just critical thinking - and when it comes to time series analysis and forecasting (TSAF) in SAS, it's all about "rethinking the curve". At the Canada Revenue Agency (CRA) Compliance Programs Branch (CPB), we have grappled with reliable forecasting for macro-level tax variables on a month-to-month basis, even before the COVID-19 pandemic hit. But now we face a particularly difficult challenge. As with many large organizations, it is not easy to foretell what the fallout may be from such a cataclysm. In setting up SAS to right the trajectory, we must be extra cautious about some of the fallacies in applying TSAF in this context: the lagged effect for tax revenues realized based on audits of the previous tax year, the need to differentiate average tax recovery per case from sum of tax recovery (month-to-month), realizing that industry sectors are not "one size fits all", and accounting for relatively temporary effects of staffing reorientation in the conversion to a virtual workplace versus the more enduring effects of business disruptions. With SAS Enterprise Miner's abilities to continuously adjust forecasts, sub-categorize datapoints by tax office or industry sector, and apply lagged cross-correlation analysis, we are suitably equipped with the right tools and this can provide abstract learnings for other large organizations. | time series analysis, prediction interval, data mining, correlation, significance level, calendar year, series plot | source presentation |
22 |
| 72 | Scaffolding SAS Projects With NPM and SASjs | Allan Bowe | Is your SAS scattered across SAS folders, local directories and server file systems? Are your deliverables delayed by the need to build tools? Does your deployment process involve manual steps? The SASjs framework enables code consistency across teams and projects, de-risks the use of shared tools and dependencies, and facilitates continuous deployment to SAS environments. The framework can be applied to SAS Viya jobs, SAS 9 Stored Processes, and even regular SAS programs on a file system. Join this session and learn how to create (scaffold) a SASjs project, add a job, add a macro dependency, add a program (include) dependency, add job init and term files, deploy the jobs to SAS, and run them as part of a flow. I will share the secret of how to deploy to SAS Viya without a client and secret. And, the entire demo will be performed from a local text editor (VSCode) | source | 14 | |
| 73 | Scalable and secure cloud infrastructure from your idea to reality delivered in hours, not months | Ivan Gomez | Today, clients expect their solution environments are rapidly up, available and elastic. Implementation time shouldn't be extended by months due to purchase and deployment of hardware. Done correctly and efficiently, environments should be available in hours or days and should be able to adjust and meet different capacity requirements in near real time. This presentation covers the processes and methodologies Zencos uses when working with our hosted solutions. The goal is to show, at a high level, how we take out time-consuming processes and replace them with automated services | machine learning, access control | source | 8 |
| 74 | Spatial Analytics With SAS: Examining Contributions to OpenStreetMap for the Covid-19 Response | Michael Matthews | Base SAS software includes powerful tools for spatial analytics that can be used in a variety of circumstances. This case study examines the contributions made by the Humanitarian OpenStreetMap Team (HOT) volunteers to support diverse global projects assisting impacted communities. SAS ODS Graphics procedures were used to analyze the OpenStreetMap data. As a result, work undertaken by global volunteers and mapping communities to identify and update infrastructure and housing can be analyzed to show the impact both spatially and over time. This analysis demonstrates techniques that are applicable for any organization using the open data from OpenStreetMap or other mapping data sources |
proc step, sas base, proc surveyreg, dataset, proc sql | source xml |
16 |
| 75 | Start/Stop SAS LASR Analytic Server using PROC HTTP | Anand Vyas | As more and more organizations have started to adopt DevOps tools and practices, automating systems and services is the norm to achieve accuracy, reliability and avoid any human errors. Many organizations have integrated their SAS platform with these tool sets to achieve maximum automation. In this paper, we will discuss how we can automate start/stop of SAS LASR Analytic serv er by making API calls using PROC HTTP |
proc http, proc step | source | 5 |
| 76 | Stretch Your Mind And Your SAS Grid | Diane Hatcher | Deploying the SAS grid in a cloud provides the ability to use cloud capabilities to automatically scale your grid based on user demand. Grid nodes can be added as user volume increases and then scaled back down when the demand decreases. This paper will discuss how this can be done across all cloud vendors, including Azure, AWS, GCP and OpenShift. Deploying the SAS grid in the cloud can help you better manage TCO and storage costs without an upfront hardware investment. In addition, the SAS grid can also be complemented with SAS running in containers. There will also be a discussion on the future of parallel processing, where containers play a leading role | sas grid, cloud storage, data cleaning | source | 6 |
| 77 | Super Static SG Graphs | Kriss Harris | Would you like to know how to utilize ODS objects to produce graphs just the way you want them? Such as with coloring and annotation? This submission will show you how to create the graphs you want through using a Kaplan- Meier example |
kaplan-meier plot, hazard ratio, macro variable, median survival, proc step | source presentation |
7 |
| 78 | Survey Data Analysis Made Easy With SAS | Melanie Dove, Katherine Heck |
Population-based, representative surveys often incorporate complex methods in data collection, such as oversampling, weighting, stratification, or clustering. If survey procedures are not used in data analyses, results will provide incorrect estimates and may overstate the results of significance testing. SAS procedures, such as PROC SURVEYFREQ and PROC SURVEYMEANS, make it easy to adjust for the complex sample design and weighting of representative surveys to obtain the correct percentages, confidence intervals, means, odds ratios, and other statistics from complex survey data. In this introductory presentation, attendees will learn why it is necessary to use survey procedures when analyzing stratified or cluster sample surveys, what key survey design features are incorporated in the SAS code, and how to generate estimates using SAS survey procedures. This presentation will provide sample survey procedure code, and explain how to interpret the output from each survey procedure |
proc step, sample, population, proc surveyfreq, proc surveymeans, stratified sampling, proc surveylogistic, proc surveyreg, random sampling, by group processing, cluster sampling, confidence interval | source presentation |
8 |
| 79 | The Beauty and Brains of Actionable Data | Joseph W. Perez | Presentation | source | 11 | |
| 80 | The Potential of Simulation Technologies: Multi-Agent Simulation and Reinforcement Learning | Satoki Fujita, Shogo Miyazawa, Ryo Kiguchi, Yuki Yoshida, Katsunari Hirano, Yoshitake Kitanishi |
The society in which we live is constantly changing, as represented by the spread of Covid - 19, which occurred in 2020 and still casts a shadow over the world. It is necessary to anticipate such changes in advance and take the best response promptly, and in such cases, various simulation techniques are useful. Among them, Multi-Agent Simulation (MAS), which is assembled from micro factors and reproduces macro phenomena by their interaction, has a high degree of freedom and can flexibly handle detailed situation settings. This paper introduces the attractiveness of it and show that by introducing Reinforcement Learning (RL) there, it is possible to search for optimal decision-making based on the phenomena reproduced by MAS. The future is predicted through simulations (MAS etc.), and the optimum response policy is derived by RL. As a first step to realize such a flow in actual tasks, we show the potential of these technologies through an example of virtual infectious disease spread | confidence interval, machine learning | source presentation |
14 |
| 81 | The SAS Data Set Characterization Utility | Michael A. Raithel | Most SAS programmers reach for two tools when they first receive a new SAS data set: PROC CONTENTS and PROC MEANS. They use PROC CONTENTS to review the data set's metadata; the physical attributes of the variables such as name, label, type, and length. They use PROC MEANS to determine the basic arithmetic characteristics of the numerical variables, such as the minimum, maximum, and mean values. Doing this involves running two different SAS procedures, combing through two separate SAS-generated reports, and correlating the information about specific variables between the disparate reports. The SAS Data Set Characterization Utility generates a single report file that contains the best of both the CONTENTS and the MEANS procedures. It produces an EXCEL file with a single row of consolidated metrics for each variable found in the SAS data set. The variable's metrics include its key metadata attributes and-for numeric variables-its basic statistical properties |
dataset, numeric variable, character variable, missing value, proc step, macro variable, excel, sas programmer, proc contents | source | 10 |
| 82 | Tips for Providing Effective and Budget-Friendly SAS Development in a Remote Environment | Kelly D. Smith | Did you know that less than one-third of employees report being engaged in their work and their organization? Did you know that appropriate professional development activities can keep employees engaged and motivated, improving both performance and retention? This presentation, suitable for all levels of SAS users, will explore ways to identify and utilize budget friendly development opportunities. Learn how to develop SAS skills while strengthening teamwork. Take away tips, links, and ideas to maximize employee engagement and learning, and make the most of your department's contributions to organizational excellence | data science, problem solving | source presentation |
9 |
| 83 | Transcoding: Understand, Troubleshoot, and Resolve a Most Mysterious SAS Error | Yun (Julie) Zhuo | In today's integrated and interconnected world, sharing SAS data among organizations and countries has become increasingly common. The transcoding error, which may occur during data transfers between incompatible SAS encoding environments, remains one of the most mysterious SAS errors to programmers. This presentation discusses three steps to demystify the transcoding issues. First, encoding, transcoding and other background information will be explained. Second, SAS procedures, options and other techniques that are useful for troubleshooting will be given. Finally, the character variable padding (CVP) engine and SAS configuration files are provided as solutions to the problem. By the end of the presentation, the audience should have a clear vision for troubleshooting and solving transcoding errors | character variable, error message, raw data, sas log, proc options, proc step, data step, proc contents, length statement | source presentation |
14 |
| 84 | Twenty Ways to Run Your SAS Programs Faster and Use Less Space | Stephen B. Sloan | When we run SAS programs that use large amounts of data or have complicated algorithms, we often are frustrated by the amount of time it takes for the programs to run and by the large amount of space required for the program to run to completion. Even experienced SAS programmers sometimes run into this situation, perhaps through the need to produce results quickly, through a change in the data source, through inheriting someone else's programs, or for some other reason. This paper outlines twenty techniques that can reduce the time and space required for a program without requiring an extended period of time for the modifications. The twenty techniques are a mixture of space-saving and time- saving techniques, and many are a combination of the two approaches. They do not require advanced knowledge of SAS, only a reasonable familiarity with Base SAS and a willingness to delve into the details of the programs | proc step, dataset, proc sort, numeric variable, proc means, by group processing, character variable | source | 4 |
| 85 | Understanding the Effects of Campus Safety on College Student Retention: A Panel Data Analysis | Zhong Zheng, James Farley, Maham Khan |
Student success can be defined positive student outcomes based on the operational activities and support provided by a learning institution. Student success outcomes measurements primarily focus on advancement towards persistence, graduation, and good academic standing but there are non-academic experiences that can also impact student success. This paper explores the relationship between campus safety and student outcomes pertaining to retention and completion rates. One hundred and thirty public and private Research Tier 1 (R1) universities were used in this analysis. The R1 universities meet the highest benchmarks in research activity and expenditures as measured by the Carnegie Classification of Institutions of Higher Education. In developing analysis, data were collected from the College Scorecard database and the official website of Campus Safety and Security. The data was then curated and analyzed using different statistical procedures in SAS Enterprise Guide and SAS Viya | correlation, regression analysis | source presentation |
7 |
| 86 | Uniform Hashing of Arbitrary Input Into Key-Exclusive Segments | Paul M. Dorfman, Donald J. Henderson |
Aggregating or combining large data volumes can challenge computing resources. For example, the process may be hindered by the system limits on utility space or memory and, as a result, either fail or run too long to be useful. It is a natural inclination to try solving the problem by segregating the input records into a number of smaller segments, processing them independently and combining the results. However, in order for such a divide-and-conquer tactic to work, two seemingly contradictory criteria must be met: First, to aggregate or combine the data correctly, no segment can share its key values with the rest; and second, the segments must be more or less equal in size. In this presentation, we show how a hash function can be used to achieve it for arbitrary input with no prior knowledge of the distribution of the key values among its records | composite key, create table, proc sql, sample, hash table, proc step, data wrangling, where statement, proc freq, hashing function | source presentation |
17 |
| 87 | Using Behavioral Insights to Increase Fine Payments | Helen Aki, Olivia Wills |
Behavioral science research and experience shows us that people do not always respond rationally to our incentives, information, and legislation. People are often influenced by seemingly unimportant factors or biases and can behave in ways that are against their best interests and even against their own intentions. One of the challenging behaviors for the Ministry of Justice has been people failing to pay their fines. The Ministry of Justice collects court-imposed fines, reparations to victims, and infringements on behalf of local authorities and the police. Over half a million New Zealanders owe fines and total collections debt is around US$395m, or US$78 for every person in New Zealand. Not paying also impacts the people who owe fines, as they can be hit with enforcement actions or may even be summoned to Court. We have used evidence from behavioral science to test new approaches to improve fines payment behavior | randomized controlled trial, debt collection | source | 9 |
| 88 | Using the %SCAN Function | Jonas V. Bilenas | Presentation: In this presentation, we will look at the MACRO function %SCAN. It is similar to the SCAN function in the DATA STEP, but is useful in macro code. We will look at two examples that use the %SCAN function. The first example will show you how to set up a macro that forces all data set variables to be capitalized. The second example will show how to split SGSCATTER plots into multiple plots |
macro variable, character functions, macro function, proc step, macro loop, proc contents, do loop, proc sgscatter, response variable, macro definition, proc sql | source presentation |
10 |
| 89 | Weight of Evidence, Dummy Variables and Degrees of Freedom | Bruce S. Lund | Predictive models with a binary target are often fitted by logistic regression. An important step in using logistic regression is transforming of predictors before the model fitting stage. In credit risk modeling and direct marketing modeling, predictors are often transformed by weight of evidence (WOE) coding. This approach applies to discrete predictors, whether nominal, ordinal, or numeric. It also applies to continuous numeric predictors after such a predictor has been reduced to discrete ranges ("fine classing"). A widely used alternative to WOE coding is dummy variable coding. Let C be a discrete predictor and C_woe be its WOE coding. A model where C_woe is the only predictor has the same probabilities as the model with C appearing in CLASS C as the only predictor. Hence, the degrees of freedom of C_woe are L-1 where C has L levels | degrees of freedom, logistic distribution, proc logistic, proc step, binary variable, logistic regression, likelihood function, forward selection, sample, credit risk analyst, proc hplogistic, credit risk, maximum likelihood estimation, macro definition, model training, sample size, regression analysis | source sas presentation |
15 |
| 90 | What Kind of WHICH Do You CHOOSE to be? | Richann J. Watson, Louise S. Hadden |
A typical task for a SAS practitioner is the creation of a new variable that is based on the value of another variable or string. This task is frequently accomplished by the use of IF- THEN-ELSE statements. However, manually typing a series of IF-THEN-ELSE statements can be time-consuming and tedious, as well as prone to typos or cut and paste errors. Serendipitously, SAS has provided us with an easier way to assign values to a new variable. The WHICH and CHOOSE functions provide a convenient and efficient method for data- driven variable creation | choose, if-then-else | source | 7 |
This site serves as a convenient hosting location and does not work as a publisher. All copyrights belong to their respective owners.
The information on this site is provided for general reference purposes only. Whilst every effort is made to ensure that the site is up to date and accurate, no warrant, nor any responsibility or liability for, the accuracy or completeness of the content or for any loss which may arise from reliance on information contained in this site is provided. Links to other sites are provided for your convenience but no responsibility or liability for the content of those sites or of any external site which links to this site is accepted. References to any specific commercial product, process, service, or corporation are also for information only, and do not constitute endorsement, recommendation or favoring.
SAS is a registered trademark of SAS Institute Inc. SAS, Statistical Analysis System, and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute Inc. in the USA and other countries. ® indicates USA registration.