Two of the powerful features of ODS Graphics procedures is the ability to create forest plots and add inner margin tables to graphics output. The drawback, however, is that the syntax required by the programmer from PROC TEMPLATE is complex and tedious. A prompted application or even a parameterized stored process that connects PROC TEMPLATE code to a point-and-click application definitely makes life easier for coders in many industries who frequently create these types of graphic output.

Read the paper (PDF) | Download the data file (ZIP)

Although the statistical foundations of predictive analytics have large overlaps, the business objectives, data availability, and regulations are different across the property and casualty insurance, life insurance, banking, pharmaceutical, and genetics industries. A common process in property and casualty insurance companies with large data sets is introduced, including data acquisition, data preparation, variable creation, variable selection, model building (also known as fitting), model validation, and model testing. Variable selection and model validation stages are described in more detail. Some successful models in the insurance companies are introduced. Base SAS^®, SAS^® Enterprise Guide^®, and SAS^® Enterprise Miner™ are presented as the main tools for this process.

The new and highly anticipated SAS^® Output Delivery System (ODS) destination for Microsoft Excel is finally here! Available as a production feature in the third maintenance release of SAS^® 9.4 (TS1M3), this new destination generates native Excel (XLSX) files that are compatible with Microsoft Office 2010 or later. This paper is written for anyone, from entry-level programmers to business analysts, who uses the SAS^® System and Microsoft Excel to create reports. The discussion covers features and benefits of the new Excel destination, differences between the Excel destination and the older ExcelXP tagset, and functionality that exists in the ExcelXP tagset that is not available in the Excel destination. These topics are all illustrated with meaningful examples. The paper also explains how you can bridge the gap that exists as a result of differences in the functionality between the destination and the tagset. In addition, the discussion outlines when it is beneficial for you to use the Excel destination versus the ExcelXP tagset, and vice versa. After reading this paper, you should be able to make an informed decision about which tool best meets your needs.

Read the paper (PDF) | Watch the recording

Geographically Weighted Negative Binomial Regression (GWNBR) was developed by Silva and Rodrigues (2014). It is a generalization of the Geographically Weighted Poisson Regression (GWPR) proposed by Nakaya and others (2005) and of the Poisson and negative binomial regressions. This paper shows a SAS^® macro to estimate the GWNBR model encoded in SAS/IML^® software and shows how to use the SAS procedure GMAP to draw the maps.

Read the paper (PDF) | Download the data file (ZIP)

For far too long, anti-money laundering and terrorist financing solutions have forced analysts to wade through oceans of transactions and alerted work items (alerts). Alert-centered analysis is both ineffective and costly. The goal of an anti-money laundering program is to reduce risk for your financial institution, and to do this most effectively, you must start with analysis at the customer level, rather than simply troll through volumes of alerts and transactions. In this session, discover how a customer-centric approach leads to increased analyst efficiency and streamlined investigations. Rather than starting with alerts and transactions, starting with a customer-centric view allows your analysts to rapidly triage suspicious activities, prioritize work, and quickly move into investigating the highest risk customer activities.

Bayesian inference for complex hierarchical models with smoothing splines is typically intractable, requiring approximate inference methods for use in practice. Markov Chain Monte Carlo (MCMC) is the standard method for generating samples from the posterior distribution. However, for large or complex models, MCMC can be computationally intensive, or even infeasible. Mean Field Variational Bayes (MFVB) is a fast deterministic alternative to MCMC. It provides an approximating distribution that has minimum Kullback-Leibler distance to the posterior. Unlike MCMC, MFVB efficiently scales to arbitrarily large and complex models. We derive MFVB algorithms for Gaussian semiparametric multilevel models and implement them in SAS/IML^® software. To improve speed and memory efficiency, we use block decomposition to streamline the estimation of the large sparse covariance matrix. Through a series of simulations and real data examples, we demonstrate that the inference obtained from MFVB is comparable to that of PROC MCMC. We also provide practical demonstrations of how to estimate additional posterior quantities of interest from MFVB either directly or via Monte Carlo simulation.

Read the paper (PDF) | Download the data file (ZIP)

The surge of data and data sources in marketing has created an analytical bottleneck in most organizations. Analytics departments have been pushed into a difficult decision: either purchase black-box analytical tools to generate efficiencies or hire more analysts, modelers, and data scientists. Knowledge gaps stemming from restrictions in black-box tools or from backlogs in the work of analytical teams have resulted in lost business opportunities. Existing big data analytics tools respond well when dealing with large record counts and small variable counts, but they fall short in bringing efficiencies when dealing with wide data. This paper discusses the importance of an agile modeling engine designed to deliver productivity, irrespective of the size of the data or the complexity of the modeling approach.

Read the paper (PDF) | Watch the recording

Even though marketing is inevitable in every business, every year the marketing budget is limited and prudent fund allocations are required to optimize marketing investment. In many businesses, the marketing fund is allocated based on the marketing manager's experience, departmental budget allocation rules, and sometimes 'gut feelings' of business leaders. Those traditional ways of budget allocation yield suboptimal results and in many cases lead to money being wasted on irrelevant marketing efforts. Marketing mixed models can be used to understand the effects of marketing activities and identify the key marketing efforts that drive the most sales among a group of competing marketing activities. The results can be used in marketing budget allocation to take out the guesswork that typically goes into the budget allocation. In this paper, we illustrate practical methods for developing and implementing marketing mixed modeling using SAS^® procedures. Real-life challenges of marketing mixed model development and execution are discussed, and several recommendations are provided to overcome some of those challenges.

Graphs are essential for many clinical and health care domains, including analysis of clinical trials safety data and analysis of the efficacy of the treatment, such as change in tumor size. Creating such graphs is a breeze with procedures from SAS^® 9.4 ODS Graphics. This paper shows how to create many industry-standard graphs such as Lipid Profile, Swimmer Plot, Survival Plot, Forest Plot with Subgroups, Waterfall Plot, and Patient Profile using Study Data Tabulation Model (SDTM) data with just a few lines of code.

Read the paper (PDF) | Download the data file (ZIP) | Watch the recording

The HIPAA Privacy Rule can restrict geographic and demographic data used in health-care analytics. After reviewing the HIPAA requirements for de-identification of health-care data used in research, this poster guides the beginning SAS^® Visual Analytics user through different options to create a better user experience. This poster presents a variety of data visualizations the analyst will encounter when describing a health-care population. We explore the different options SAS Visual Analytics offers and also offer tips on data preparation prior to using SAS^® Visual Analytics Designer. Among the topics we cover are SAS Visual Analytics Designer object options (including geo bubble map, geo region map, crosstab, and treemap), tips for preparing your data for use in SAS Visual Analytics, and tips on filtering data after it's been loaded into SAS Visual Analytics, and more.

SAS^® Grid Manager, as well as other grid computing technologies, have a set of great capabilities that we, IT professionals, love to have in our systems. This technology increases high availability, allows parallel processing, facilitates increasing demand by scale out, and offers other features that make life better for those managing and using these environments. However, even when business users take advantage of these features, they are more concerned about the business part of the problem. Most of the time business groups hold the budgets and are key stakeholders for any SAS Grid Manager project. Therefore, it is crucial to demonstrate to business users how they will benefit from the new technologies, how the features will improve their daily operations, help them be more efficient and productive, and help them achieve better results. This paper guides you through a process to create a strong and persuasive business plan that translates the technology features from SAS Grid Manager to business benefits.

Read the paper (PDF) | Watch the recording

There are standard risk metrics financial institutions use to assess the risk of a portfolio. These include well known measures like value at risk and expected shortfall and related measures like contribution value at risk. While there are industry-standard approaches for calculating these measures, it is often the case that financial institutions have their own methodologies. Further, financial institutions write their own measures, in addition to the common risk measures. SAS^® High-Performance Risk comes equipped with over 20 risk measures that use standard methodology, but the product also allows customers to define their own risk measures. These user-defined statistics are treated the same way as the built-in measures, but the logic is specified by the customer. This paper leads the user through the creation of custom risk metrics using the HPRISK procedure.

As SAS^® programmers, we often develop listings, graphs, and reports that need to be delivered frequently to our customers. We might decide to manually run the program every time we get a request, or we might easily schedule an automatic task to send a report at a specific date and time. Both scenarios have some disadvantages. If the report is manual, we have to find and run the program every time someone request an updated version of the output. It takes some time and it is not the most interesting part of the job. If we schedule an automatic task in Windows, we still sometimes get an email from the customers because they need the report immediately. That means that we have to find and run the program for them. This paper explains how we developed an on-demand report platform using SAS^® Enterprise Guide^®, SAS^® Web Application Server, and stored processes. We had developed many reports for different customer groups, and we were getting more and more emails from them asking for updated versions of their reports. We felt we were not using our time wisely and decided to create an infrastructure where users could easily run their programs through a web interface. The tool that we created enables SAS programmers to easily release on-demand web reports with minimum programming. It has web interfaces developed using stored processes for the administrative tasks, and it also automatically customizes the front end based on the user who connects to the website. One of the challenges of the project was that certain reports had to be available to a specific group of users only.

You have SAS^® Enterprise Guide^® installed. You use SAS Enterprise Guide in your day-to-day work. You see how Enterprise Guide can be an aid to accessing data and insightful analytics. You have people you work with or support who are new to SAS^® and want to learn. You have people you work with or support who don't particularly want to code but use the GUI and wizard within Enterprise Guide. And then you have the spreadsheet addict, the person or group who refuse to even sign on to SAS. These people need to consume the data sitting in SAS, and they need to do analysis, but they want to do it all in a spreadsheet. But you need to retain an audit trail of the data, and you have to reduce the operational risk of using spreadsheets for reporting. What do you do? This paper shares some of the challenges and triumphs in empowering these very different groups of people using SAS.

As Data Management professionals, you have to comply with new regulations and controls. One such regulation is Basel Committee on Banking Supervision (BCBS) 239. To respond to these new demands, you have to put processes and methods in place to automate metadata collection and analysis, and to provide rigorous documentation around your data flows. You also have to deal with many aspects of data management including data access, data manipulation (ETL and other), data quality, data usage, and data consumption, often from a variety of toolsets that are not necessarily from a single vendor. This paper shows you how to use SAS^® technologies to support data governance requirements, including third party metadata collection and data monitoring. It highlights best practices such as implementing a business glossary and establishing controls for monitoring data. Attend this session to become familiar with the SAS tools used to meet the new requirements and to implement a more managed environment.

Generalized linear models (GLMs) are commonly used to model rating factors in insurance pricing. The integration of territory rating and geospatial variables poses a unique challenge to the traditional GLM approach. Generalized additive models (GAMs) offer a flexible alternative based on GLM principles with a relaxation of the linear assumption. We explore two approaches for incorporating geospatial data in a pricing model using a GAM-based framework. The ability to incorporate new geospatial data and improve traditional approaches to territory ratemaking results in further market segmentation and a better match of price to risk. Our discussion highlights the use of the high-performance GAMPL procedure, which is new in SAS/STAT^® 14.1 software. With PROC GAMPL, we can incorporate the geographic effects of geospatial variables on target loss outcomes. We illustrate two approaches. In our first approach, we begin by modeling the predictors as regressors in a GLM, and subsequently model the residuals as part of a GAM based on location coordinates. In our second approach, we model all inputs as covariates within a single GAM. Our discussion compares the two approaches and demonstrates visualization of model outputs.

SAS^® Embedded Process offers a flexible, efficient way to leverage increasing amounts of data by injecting the processing power of SAS^® directly where the data lives. SAS Embedded Process can tap into the massively parallel processing (MPP) architecture of Hadoop for scalable performance. Using SAS^® In-Database Technologies for Hadoop, you can run scoring models generated by SAS^® Enterprise Miner™ or, with SAS^® In-Database Code Accelerator for Hadoop, user-written DS2 programs in parallel. With SAS Embedded Process on Hadoop you can also perform data quality operations, and extract and transform data using SAS^® Data Loader. This paper explores key SAS technologies that run inside the Hadoop parallel processing framework and prepares you to get started with them.

Injury severity describes the severity of the injury to the person involved in the crash. Understanding the factors that influence injury severity can be helpful in designing mechanisms to reduce accident fatalities. In this research, we model and analyze the data as a hierarchy with three levels to answer the question what road, vehicle and driver-related factors influence injury severity. In this study, we used hierarchical linear modeling (HLM) for analyzing nested data from Fatality Analysis Reporting System (FARS). The results show that driver-related factors are directly related to injury severity. On the other hand, road conditions and vehicle characteristics have significant moderation impact on injury severity. We believe that our study has important policy implications for designing customized mechanisms specific to each hierarchical level to reduce the occurrence of fatal accidents.

Do you create complex reports using PROC REPORT? Are you confused by the COMPUTE BLOCK feature of PROC REPORT? Are you even aware of it? Maybe you already produce reports using PROC REPORT, but suddenly your boss needs you to modify some of the values in one or more of the columns. Maybe your boss needs to see the values of some rows in boldface and others highlighted in a stylish yellow. Perhaps one of the columns in the report needs to display a variety of fashionable formats (some with varying decimal places and some without any decimals). Maybe the customer needs to see a footnote in specific cells of the report. Well, if this sounds familiar then come take a look at the COMPUTE BLOCK of PROC REPORT. This paper shows a few tips and tricks of using the COMPUTE DEFINE block with conditional IF/THEN logic to make your reports stylish and fashionable. The COMPUTE BLOCK allows you to use data DATA step code within PROC REPORT to provide customization and style to your reports. We'll see how the Census Bureau produces a stylish demographic profile for customers of its Special Census program using PROC REPORT with the COMPUTE BLOCK. The paper focuses on how to use the COMPUTE BLOCK to create this stylish Special Census profile. The paper shows quick tips and simple code to handle multiple formats within the same column, make the values in the Total rows boldface, trafficlighting, and how to add footnotes to any cell based on the column or row. The Special Census profile report is an Excel table created with ODS tagsets.ExcelXP that is stylish and fashionable, thanks in part to the COMPUTE BLOCK.

Read the paper (PDF) | Watch the recording

Hierarchical nonlinear mixed models are complex models that occur naturally in many fields. The NLMIXED procedure's ability to fit linear or nonlinear models with standard or general distributions enables you to fit a wide range of such models. SAS/STAT^® 13.2 enhanced PROC NLMIXED to support multiple RANDOM statements, enabling you to fit nested multilevel mixed models. This paper uses an example to illustrate the new functionality.

We introduce age-period-cohort (APC) models, which analyze data in which performance is measured by age of an account, account open date, and performance date. We demonstrate this flexible technique with an example from a recent study that seeks to explain the root causes of the US mortgage crisis. In addition, we show how APC models can predict website usage, retail store sales, salesperson performance, and employee attrition. We even present an example in which APC was applied to a database of tree rings to reveal climate variation in the southwestern United States.

If your organization already deploys one or more software solutions via Amazon Web Services (AWS), you know the value of the public cloud. AWS provides a scalable public cloud with a global footprint, allowing users access to enterprise software solutions anywhere at any time. Although SAS^® began long before AWS was even imagined, many loyal organizations driven by SAS are moving their local SAS analytics into the public AWS cloud, alongside other software hosted by AWS. SAS^® Solutions OnDemand has assisted organizations in this transition. In this paper, we describe how we extended our enterprise hosting business to AWS. We describe the open source automation framework from which SAS Soultions onDemand built our automation stack, which simplified the process of migrating a SAS implementation. We'll provide the technical details of our automation and network footprint, a discussion of the technologies we chose along the way, and a list of lessons learned.

Project management is a hot topic across many industries, and there are multiple commercial software applications for managing projects available. The reality, however, is that the majority of project management software is not applicable for daily usage. SAS^® has a solution for this issue that can be used for managing projects graphically in real time. This paper introduces a new paradigm for project management using the SAS^® Graph Template Language (GTL). SAS clients, in real time, can use GTL to visualize resource assignments, task plans, delivery tracking, and project status across multiple project levels for more efficient project management.

Since Atul Gawande popularized the term in describing the work of Dr. Jeffrey Brenner in a New Yorker article, hot-spotting has been used in health care to describe the process of identifying super-utilizers of health care services, then defining intervention programs to coordinate and improve their care. According to Brenner's data from Camden, New Jersey, 1% of patients generate 30% of payments to hospitals, while 5% of patients generate 50% of payments. Analyzing administrative health care claims data, which contains information about diagnoses, treatments, costs, charges, and patient sociodemographic data, can be a useful way to identify super-users, as well as those who may be receiving inappropriate care. Both groups can be targeted for care management interventions. In this paper, techniques for patient outlier identification and prioritization are discussed using examples from private commercial and public health insurance claims data. The paper also describes techniques used with health care claims data to identify high-risk, high-cost patients and to generate analyses that can be used to prioritize patients for various interventions to improve their health.

Contemporary data-collection processes usually involve recording information about the geographic location of each observation. This geospatial information provides modelers with opportunities to examine how the interaction of observations affects the outcome of interest. For example, it is likely that car sales from one auto dealership might depend on sales from a nearby dealership either because the two dealerships compete for the same customers or because of some form of unobserved heterogeneity common to both dealerships. Knowledge of the size and magnitude of the positive or negative spillover effect is important for creating pricing or promotional policies. This paper describes how geospatial methods are implemented in SAS/ETS^® and illustrates some ways you can incorporate spatial data into your modeling toolkit.

Have you ever wondered how to get the most from Web 2.0 technologies in order to visualize SAS^® data? How to make those graphs dynamic, so that users can explore the data in a controlled way, without needing prior knowledge of SAS products or data science? Wonder no more! In this session, you learn how to turn basic sashelp.stocks data into a snazzy HighCharts stock chart in which a user can review any time period, zoom in and out, and export the graph as an image. All of these features with only two DATA steps and one SORT procedure, for 57 lines of SAS code.

Download the data file (ZIP) | View the e-poster or slides (PDF)

In the aftermath of the 2008 global financial crisis, banks had to improve their data risk aggregation in order to effectively identify and manage their credit exposures and credit risk, create early warning signs, and improve the ability of risk managers to challenge the business and independently assess and address evolving changes in credit risk. My presentation focuses on using SAS^® Credit Risk Dashboard to achieve all of the above. Clearly, you can use my method and principles of building a credit risk dashboard to build other dashboards for other types of risks as well (market, operational, liquidity, compliance, reputation, etc.). In addition, because every bank must integrate the various risks with a holistic view, each of the risk dashboards can be the foundation for building an effective enterprise risk management (ERM) dashboard that takes into account correlation of risks, risk tolerance, risk appetite, breaches of limits, capital allocation, risk-adjusted return on capital (RAROC), and so on. This will support the actions of top management so that the bank can meet shareholder expectations in the long term.

Looking for new ways to improve your business? Try mining your own data! Event log data is a side product of information systems generated for audit and security purposes and is seldom analyzed, especially in combination with business data. Along with the cloud computing era, more event log data has been accumulated and analysts are searching for innovative ways to take advantage of all data resources in order to get valuable insights. Process mining, a new field for discovering business patterns from event log data, has recently proved useful for business applications. Process mining shares some algorithms with data mining but it is more focused on interpretation of the detected patterns rather than prediction. Analysis of these patterns can lead to improvements in the efficiency of common existing and planned business processes. Through process mining, analysts can uncover hidden relationships between resources and activities and make changes to improve organizational structure. This paper shows you how to use SAS^® Analytics to gain insights from real event log data.

Working with big data is often time consuming and challenging. The primary goal in programming is to maximize throughputs while minimizing the use of computer processing time, real time, and programmers' time. By using the Multiprocessing (MP) CONNECT method on a symmetric multiprocessing (SMP) computer, a programmer can divide a job into independent tasks and execute the tasks as threads in parallel on several processors. This paper demonstrates the development and application of a parallel processing program on a large amount of health-care data.

Read the paper (PDF) | View the e-poster or slides (PDF)

Real-time, integrated marketing solutions are a necessity for maintaining your competitive advantage. This presentation provides a brief overview of three SAS products (SAS^® Marketing Automation, SAS^® Real-Time Decision Manager, and SAS^® Event Stream Processing) that form a basis for building modern, real-time, interactive marketing solutions. It presents typical (and also possible) customer-use cases that you can implement with a comprehensive real-time interactive marketing solution, in major industries like finance (banking), telco, and retail. It demonstrates typical functional architectures that need to be implemented to support business cases (how solution components collaborate with customer's IT landscape and with each other). And it provides examples of our experience in implementing these solutions--dos and don'ts, best practices, and what to expect from an implementation project.

Read the paper (PDF) | Watch the recording

Microsoft Visual Basic Scripting Edition (VBScript) and SAS^® software are each powerful tools in their own right. These two technologies can be combined so that SAS code can call a VBScript program or vice versa. This gives a programmer the ability to automate SAS tasks; traverse the file system; send emails programmatically via Microsoft Outlook or SMTP; manipulate Microsoft Word, Microsoft Excel, and Microsoft PowerPoint files; get web data; and more. This paper presents example code to demonstrate each of these capabilities.

Read the paper (PDF) | Download the data file (ZIP)

Do you want to see and experience how to configure SAS^® Enterprise Miner™ single sign-on? Are you looking to explore setting up Integrated Windows Authentication with SAS^® Visual Analytics? This hands-on workshop demonstrates how you can configure Kerberos delegation with SAS^® 9.4. You see how to validate the prerequisites, make the configuration changes, and use the applications. By the end of this workshop you will be empowered to start your own configuration.

Considering the fact that SAS^® Grid Manager is becoming more and more popular, it is important to fulfill the user's need for a successful migration to a SAS^® Grid environment. This paper focuses on key requirements and common issues for new SAS Grid users, especially if they are coming from a traditional environment. This paper describes a few common requirements like the need for a current working directory, the change of file system navigation in SAS^® Enterprise Guide^® with user-given location, getting job execution summary email, and so on. The GRIDWORK directory has been introduced in SAS Grid Manager, which is a bit different from the traditional SAS WORK location. This paper explains how you can use the GRIDWORK location in a more user-friendly way. Sometimes users experience data set size differences during grid migration. A few important reasons for data set size difference are demonstrated. We also demonstrate how to create new custom scripts as per business needs and how to incorporate them with SAS Grid Manager engine.

Read the paper (PDF) | View the e-poster or slides (PDF)

From stock price histories to hospital stay records, analysis of time series data often requires use of lagged (and occasionally lead) values of one or more analysis variables. For the SAS^® user, the central operational task is typically getting lagged (lead) values for each time point in the data set. While SAS has long provided a LAG function, it has no analogous lead function--an especially significant problem in the case of large data series. This paper reviews the LAG function, in particular the powerful, but non-intuitive implications of its queue-oriented basis. The paper demonstrates efficient ways to generate leads with the same flexibility as the LAG function, but without the common and expensive recourse to data re-sorting. It also shows how to dynamically generate leads and lags through use of the hash object.

Is uniqueness essential for your reports? SAS^® Visual Analytics provides the ability to customize your reports to make them unique by using the SAS^® Theme Designer. The SAS Theme Designer can be accessed from the SAS^® Visual Analytics Hub to create custom themes to meet your branding needs and to ensure a unified look across your company. The report themes affect the colors, fonts, and other elements that are used in tables and graphs. The paper explores how to access SAS Theme Designer from the SAS Visual Analytics home page, how to create and modify report themes that are used in SAS Visual Analytics, how to create report themes from imported custom themes, and how to import and export custom report themes.

Business Intelligence users analyze business data in a variety of ways. Seventy percent of business data contains location information. For in-depth analysis, it is essential to combine location information with mapping. New analytical capabilities are added to SAS^® Visual Analytics, leveraging the new partnership with Esri, a leader in location intelligence and mapping. The new capabilities enable users to enhance the analytical insights from SAS Visual Analytics. This paper demonstrates and discusses the new partnership with Esri and the new capabilities added to SAS Visual Analytics.

For SAS^® Enterprise Guide^® users, sometimes macro variables and their values need to be brought over to the local workspace from the server, especially when multiple data sets or outputs need to be written to separate files in a local drive. Manually retyping the macro variables and their values in the local workspace after they have been created on the server workspace would be time-consuming and error-prone, especially when we have quite a number of macro variables and values to bring over. Instead, this task can be achieved in an efficient manner by using dictionary tables and the CALL SYMPUT routine, as illustrated in more detail below. The same approach can also be used to bring macro variables and their values from the local to the server workspace.

Read the paper (PDF) | Download the data file (ZIP) | Watch the recording

SAS^® Visual Analytics Explorer puts the robust power of decision trees at your fingertips, enabling you to visualize and explore how data is structured. Decision trees help analysts better understand discrete relationships within data by visually showing how combinations of variables lead to a target indicator. This paper explores the practical use of decision trees in SAS Visual Analytics Explorer through an example of risk classification in the financial services industry. It explains various parameters and implications, explores ways the decision tree provides value, and provides alternative methods to help you the reality of imperfect data.

Read the paper (PDF) | Watch the recording

Business problems have become more stratified and micro-segmentation is driving the need for mass-scale, automated machine learning solutions. Additionally, deployment environments include diverse ecosystems, requiring hundreds of models to be built and deployed quickly via web services to operational systems. The new SAS^® automated modeling tool allows you to build and test hundreds of models across all of the segments in your data, testing a wide variety of machine learning techniques. The tool is completely customizable, allowing you transparent access to all modeling results. This paper shows you how to identify hundreds of champion models using SAS^® Factory Miner, while generating scoring web services using SAS^® Decision Manager. Immediate benefits include efficient model deployments, which allow you to spend more time generating insights that might reveal new opportunities, expose hidden risks, and fuel smarter, well-timed decisions.

Every day, businesses have to remain vigilant of fraudulent activity, which threatens customers, partners, employees, and financials. Normally, networks of people or groups perpetrate deviant activity. Finding these connections is now made easier for analysts with SAS^® Visual Investigator, an upcoming SAS^® solution that ultimately minimizes the loss of money and preserves mutual trust among its shareholders. SAS Visual Investigator takes advantage of the capabilities of the new SAS^® In-Memory Server. Investigators can efficiently investigate suspicious cases across business lines, which has traditionally been difficult. However, the time required to collect, process and identify emerging fraud and compliance issues has been costly. Making proactive analysis accessible to analysts is now more important than ever. SAS Visual Investigator was designed with this goal in mind and a key component is the visual social network view. This paper discusses how the network analysis view of SAS Visual Investigator, with all its dynamic visual capabilities, can make the investigative process more informative and efficient.

When analyzing data with SAS^®, we often encounter missing or null values in data. Missing values can arise from the availability, collectibility, or other issues with the data. They represent the imperfect nature of real data. Under most circumstances, we need to clean, filter, separate, impute, or investigate the missing values in data. These processes can take up a lot of time, and they are annoying. For these reasons, missing values are usually unwelcome and need to be avoided in data analysis. There are two sides to every coin, however. If we can think outside the box, we can take advantage of the negative features of missing values for positive uses. Sometimes, we can create and use missing values to achieve our particular goals in data manipulation and analysis. These approaches can make data analyses convenient and improve work efficiency for SAS programming. This kind of creative and critical thinking is the most valuable quality for data analysts. This paper exploits real-world examples to demonstrate the creative uses of missing values in data analysis and SAS programming, and discusses the advantages and disadvantages of these methods and approaches. The illustrated methods and advanced programming skills can be used in a wide variety of data analysis and business analytics fields.

You've heard all the talk about SAS^® Visual Analytics--but maybe you are still confused about how the product would work in your SAS^® environment. Many customers have the same points of confusion about what they need to do with their data, how to get data into the product, how SAS Visual Analytics would benefit them, and even should they be considering Hadoop or the cloud. In this paper, we cover the questions we are asked most often about implementation, administration, and usage of SAS Visual Analytics.

Read the paper (PDF) | Watch the recording

Inspired by Christianna William's paper on transitioning to PROC SQL from the DATA step, this paper aims to help SQL programmers transition to SAS^® by using PROC SQL. SAS adapted the Structured Query Language (SQL) by means of PROC SQL back with SAS^®6. PROC SQL syntax closely resembles SQL. However, there are some SQL features that are not available in SAS. Throughout this paper, we outline common SQL tasks and how they might differ in PROC SQL. We also introduce useful SAS features that are not available in SQL. Topics covered are appropriate for novice SAS users.

SAS^® software provides many DATA step functions that search and extract patterns from a character string, such as SUBSTR, SCAN, INDEX, TRANWRD, etc. Using these functions to perform pattern matching often requires you to use many function calls to match a character position. However, using the Perl regular expression (PRX) functions or routines in the DATA step improves pattern-matching tasks by reducing the number of function calls and making the program easier to maintain. This talk, in addition to discussing the syntax of Perl regular expressions, demonstrates many real-world applications.

Read the paper (PDF) | Download the data file (ZIP)

In recent years, many companies are trying to understand the rare events that are very critical in the current business environment. But a data set with rare events is always imbalanced and the models developed using this data set cannot predict the rare events precisely. Therefore, to overcome this issue, a data set needs to be sampled using specialized sampling techniques like over-sampling, under-sampling, or the synthetic minority over-sampling technique (SMOTE). The over-sampling technique deals with randomly duplicating minority class observations, but this technique might bias the results. The under-sampling technique deals with randomly deleting majority class observations, but this technique might lose information. SMOTE sampling deals with creating new synthetic minority observations instead of duplicating minority class observations or deleting the majority class observations. Therefore, this technique can overcome the problems, like biased results and lost information, found in other sampling techniques. In our research, we used an imbalanced data set containing results from a thyroid test with 3,163 observations, out of which only 4.7 percent of the observations had positive test results. Using SAS^® procedures like PROC SURVERYSELECT and PROC MODECLUS, we created over-sampled, under-sampled, and the SMOTE sampled data set in SAS^® Enterprise Guide^®. Then we built decision tree, gradient boosting, and rule induction models using four different data sets (non-sampled, majority under-sampled, minority over-sampled with majority under-sampled, and minority SMOTE sampled with majority under-sampled) in SAS^® Enterprise Miner™. Finally, based on the receiver operating characteristic (ROC) index, Kolmogorov-Smirnov statistics, and the misclassification rate, we found that the models built using minority SMOTE sampled with the majority under-sampled data yields better output for this data set.

In early 2006, the United States experienced a housing bubble that affected over half of the American states. It was one of the leading causes of the 2007-2008 financial recession. Primarily, the overvaluation of housing units resulted in foreclosures and prolonged unemployment during and after the recession period. The main objective of this study is to predict the current market value of a housing unit with respect to fair market rent, census region, metropolitan statistical area, area median income, household income, poverty income, number of units in the building, number of bedrooms in the unit, utility costs, other costs of the unit, and so on, to determine which factors affect the market value of the housing unit. For the purpose of this study, data was collected from the Housing Affordability Data System of the US Department of Housing and Urban Development. The data set contains 20 variables and 36,675 observations. To select the best possible input variables, several variable selection techniques were used. For example, LARS (least angle regression), LASSO (least absolute shrinkage and selection operator), adaptive LASSO, variable selection, variable clustering, stepwise regression, (PCA) principal component analysis only with numeric variables, and PCA with all variables were all tested. After selecting input variables, numerous modeling techniques were applied to predict the current market value of a housing unit. An in-depth analysis of the findings revealed that the current market value of a housing unit is significantly affected by the fair market value, insurance and other costs, structure type, household income, and more. Furthermore, a higher household income and median income of an area are associated with a higher market value of a housing unit.

Read the paper (PDF) | View the e-poster or slides (PDF)

In a data warehousing system, change data capture (CDC) plays an important part not just in making the data warehouse (DWH) aware of the change but also in providing a means of flowing the change to the DWH marts and reporting tables so that we see the current and latest version of the truth. This and slowly changing dimensions (SCD) create a cycle that runs the DWH and provides valuable insights in the history and for the decision-making future. What if the source has no CDC? It would be an ETL nightmare to identify the exact change and report the absolute truth. If these two processes can be combined into a single process where just one single transform does both jobs of identifying the change and applying the change to the DWH, then we can save significant processing times and value resources of the system. Hence, I came up with a hybrid SCD with CDC approach for this. My paper focuses on sources that DO NOT have CDC in their sources and need to perform SCD Type 2 on such records without worrying about data duplications and increased processing times.

Read the paper (PDF) | Watch the recording

Horizontal data sorting is a very useful SAS^® technique in advanced data analysis when you are using SAS programming. Two years ago (SAS^® Global Forum Paper 376-2013), we presented and illustrated various methods and approaches to perform horizontal data sorting, and we demonstrated its valuable application in strategic data reporting. However, this technique can also be used as a creative analytic method in advanced business analytics. This paper presents and discusses its innovative and insightful applications in product purchase sequence analyses such as product opening sequence analysis, product affinity analysis, next best offer analysis, time-span analysis, and so on. Compared to other analytic approaches, the horizontal data sorting technique has the distinct advantages of being straightforward, simple, and convenient to use. This technique also produces easy-to-interpret analytic results. Therefore, the technique can have a wide variety of applications in customer data analysis and business analytics fields.

Read the paper (PDF) | View the e-poster or slides (PDF)

Value-based reimbursement is the emerging strategy in the US healthcare system. The premise of value-based care is simple in concept--high quality and low cost provides the greatest value to patients and the various parties that fund their coverage. The basic equation for value is equally simple to compute: value=quality/cost. However, there are significant challenges to measuring it accurately. Error or bias in measuring value could result in the failure of this strategy to ultimately improve the healthcare system. This session discusses various methods and issues with risk adjustment in a value-based reimbursement model. Risk adjustment is an essential tool for ensuring that fair comparisons are made when deciding what health services and health providers have high value. The goal this presentation is to give analysts an overview of risk adjustment and to provide guidance for when, why, and how to use risk adjustment when quantifying performance of health services and healthcare providers on both cost and quality. Statistical modeling approaches are reviewed and practical issues with developing and implementing the models are discussed. Real-world examples are also provided.

This session is an in-depth review of SAS^® Grid performance on IBM Hardware. This review spans our environment's growth over the last four years and includes the latest upgrade to our environment from the first maintenance release of SAS^® 9.3 to the third maintenance release of SAS^® 9.4 (and doing a hardware refresh in the process).

Mobile devices are an integral part of a business professional's life. These mobile devices are getting increasingly powerful in terms of processor speeds and memory capabilities. Business users can benefit from a more analytical visualization of the data along with their business context. The new SAS^® Mobile BI contains many enhancements that facilitate the use of SAS^® Analytics in the newest version of SAS^® Visual Analytics. This paper demonstrates how to use the new analytical visualization that has been added to SAS Mobile BI from SAS Visual Analytics, for a richer and more insightful experience for business professionals on the go.

Disease prevalence is one of the most basic measures of the burden of disease in the field of epidemiology. As an estimate of the total number of cases of disease in a given population, prevalence is a standard in public health analysis. The prevalence of diseases in a given area is also frequently at the core of governmental policy decisions, charitable organization funding initiatives, and countless other aspects of everyday life. However, all too often, prevalence estimates are restricted to descriptive estimates of population characteristics when they could have a much wider application through the use of inferential statistics. As an estimate based on a sample from a population, disease prevalence can vary based on random fluctuations in that sample rather than true differences in the population characteristic. Statistical inference uses a known distribution of this sampling variation to perform hypothesis tests, calculate confidence intervals, and perform other advanced statistical methods. However, there is no agreed-upon sampling distribution of the prevalence estimate. In cases where the sampling distribution of an estimate is unknown, statisticians frequently rely on the bootstrap re-sampling procedure first given by Efron in 1979. This procedure relies on the computational power of software to generate repeated pseudo-samples similar in structure to an original, real data set. These multiple samples allow for the construction of confidence intervals and statistical tests to make statistical determinations and comparisons using the estimated prevalence. In this paper, we use the bootstrapping capabilities of SAS^® 9.4 to compare statistically the difference between two given prevalence rates. We create a bootstrap analog to the two-sample t test to compare prevalence rates from two states despite the fact that the sampling distribution of these estimates is unknown using SAS^®.

This paper discusses a set of practical recommendations for optimizing the performance and scalability of your Hadoop system using SAS^®. Topics include recommendations gleaned from actual deployments from a variety of implementations and distributions. Techniques cover tips for improving performance and working with complex Hadoop technologies such as Kerberos, techniques for improving efficiency when working with data, methods to better leverage the SAS in Hadoop components, and other recommendations. With this information, you can unlock the power of SAS in your Hadoop system.

All public schools in the United States require health and safety education for their students. Furthermore, almost all states require driver education before minors can obtain a driver's license. Through extensive analysis of the Fatality Analysis Reporting System data, we have concluded that from 2011-2013 an average of 12.1% of all individuals killed in a motor vehicle accident in the United States, District of Columbia, and Puerto Rico were minors (18 years or younger). Our goal is to offer insight within our analysis in order to better road safety education to prevent future premature deaths involving motor vehicles.

VBA has been described as a glue language, and has been widely used in exchanging data between Microsoft products such as Excel and Word or PowerPoint. How to trigger the VBA macro from SAS^® via DDE has been widely discussed in recent years. However, using SAS to send parameters to a VBA macro was seldom reported. This paper provides a solution for this problem. Copying Excel tables to PowerPoint using the combination of SAS and VBA is illustrated as an example. The SAS program rapidly scans all Excel files that are contained in one folder, passes the file information to VBA as parameters, and triggers the VBA macro to write PowerPoint files in a loop. As a result, a batch of PowerPoint files can be generated by just one mouse-click.

Read the paper (PDF) | Watch the recording

For marketers who are responsible for identifying the best customer to target in a campaign, it is often daunting to determine which media channel, offer, or campaign program is the one the customer is more apt to respond to, and therefore, is more likely to increase revenue. This presentation examines the components of designing campaigns to identify promotable segments of customers and to target the optimal customers using SAS^® Marketing Automation integrated with SAS^® Marketing Optimization.

This paper is a primer on the practice of designing, selecting, and making inferences on a statistical sample, where the goal is to estimate the magnitude of error in a book value total. Although the concepts and syntax are presented through the lens of an audit of health-care insurance claim payments, they generalize to other contexts. After presenting the fundamental measures of uncertainty that are associated with sample-based estimates, we outline a few methods to estimate the sample size necessary to achieve a targeted precision threshold. The benefits of stratification are also explained. Finally, we compare several viable estimators to quantify the book value discrepancy, making note of the scenarios where one might be preferred over the others.

Specifying colors based on group value is a quite popular practice in visualizing data, but it is not so easy to do, especially when there are multiple group values. This paper explores three different methods to dynamically assign colors to plots based on their group values. They are combining EVAL and IFN functions in the plot statements; bringing the DISCRETEATTRMAP block into the plot statements; and using the macro from the SAS^® sample 40255.

Read the paper (PDF) | Watch the recording

Are you going to enable HTTPS for your SAS^® environment? Looking to improve the security of your SAS deployment? Do you need more details about how to efficiently configure HTTPS? This paper guides you through the configuration of SAS^® 9.4 with HTTPS for the SAS middle tier. We examine how best to implement site-signed Transport Layer Security (TLS) certificates and explore how far you can take the encryption. This paper presents tips and proven practices that can help you be successful.

SAS^® High-Performance Risk distributes financial risk data and big data portfolios with complex analyses across a networked Hadoop Distributed File System (HDFS) grid to support rapid in-memory queries for hundreds of simultaneous users. This data is extremely complex and must be stored in a proprietary format to guarantee data affinity for rapid access. However, customers still desire the ability to view and process this data directly. This paper demonstrates how to use the HPRISK custom file reader to directly access risk data in Hadoop MapReduce jobs, using the HPDS2 procedure and the LASR procedure.

Read the paper (PDF) | Download the data file (ZIP)

Sensitive data requires elevated security requirements and the flexibility to apply logic that subsets data based on user privileges. Following the instructions in SAS^® Visual Analytics: Administration Guide gives you the ability to apply row-level permission conditions. After you have set the permissions, you have to prove through audits who has access and row-level security. This paper provides you with the ability to easily apply, validate, report, and audit all tables that have row-level permissions, along with the groups, users, and conditions that will be applied. Take the hours of maintenance and lack of visibility out of row-level secure data and build confidence in the data and analytics that are provided to the enterprise.

Read the paper (PDF) | Download the data file (ZIP)

For SAS^® users, PROC TABULATE and PROC REPORT (and its compute blocks) are probably among the most common procedures for calculating and displaying data. It is, however, pretty difficult to calculate and display changes from one column to another using data from other rows with just these two procedures. Compute blocks in PROC REPORT can calculate additional columns, but it would be challenging to pick up values from other rows as inputs. This presentation shows how PROC TABULATE can work with the lag(n) function to calculate rates of change from one period of time to another. This offers the flexibility of feeding into calculations the data retrieved from other rows of the report. PROC REPORT is then used to produce the desired output. The same approach can also be used in a variety of scenarios to produce customized reports.

Read the paper (PDF) | Download the data file (ZIP) | Watch the recording

The latest releases of SAS^® Data Integration Studio, SAS^® Data Management Studio and SAS^® Data Integration Server, SAS^® Data Governance, and SAS/ACCESS^® software provide a comprehensive and integrated set of capabilities for collecting, transforming, and managing your data. The latest features in the product suite include capabilities for working with data from a wide variety of environments and types including Hadoop, cloud, RDBMS, files, unstructured data, streaming, and others, and the ability to perform ETL and ELT transformations in diverse run-time environments including SAS^®, database systems, Hadoop, Spark, SAS^® Analytics, cloud, and data virtualization environments. There are also new capabilities for lineage, impact analysis, clustering, and other data governance features for enhancements to master data and support metadata management. This paper provides an overview of the latest features of the SAS^® Data Management product suite and includes use cases and examples for leveraging product capabilities.

Each night on the news we hear the level of the Dow Jones Industrial Average along with the 'first difference,' which is today's price-weighted average minus yesterday's. It is that series of first differences that excites or depresses us each night as it reflects whether stocks made or lost money that day. Furthermore, the differences form the data series that has the most addressable statistical features. In particular, the differences have the stationarity requirement, which justifies standard distributional results such as asymptotically normal distributions of parameter estimates. Differencing arises in many practical time series because they seem to have what are called 'unit roots,' which mathematically indicate the need to take differences. In 1976, Dickey and Fuller developed the first well-known tests to decide whether differencing is needed. These tests are part of the ARIMA procedure in SAS/ETS^® in addition to many other time series analysis products. I'll review a little of what is was like to do the development and the required computing back then, say a little about why this is an important issue, and focus on examples.

Read the paper (PDF) | Watch the recording

For many organizations, the answer to whether to manage their data and analytics in a public or private cloud is going to be both. Both can be the answer for many different reasons: common sense logic not to replace a system that already works just to incorporate something new; legal or corporate regulations that require some data, but not all data, to remain in place; and even a desire to provide local employees with a traditional data center experience while providing remote or international employees with cloud-based analytics easily managed through software deployed via Amazon Web Services (AWS). In this paper, we discuss some of the unique technical challenges of managing a hybrid environment, including how to monitor system performance simultaneously for two different systems that might not share the same infrastructure or even provide comparable system monitoring tools; how to manage authorization when access and permissions might be driven by two different security technologies that make implementation of a singular protocol problematic; and how to ensure overall automation of two platforms that might be independently automated, but not originally designed to work together. In this paper, we share lessons learned from a decade of experience implementing hybrid cloud environments.

Even if you're not a GIS mapping pro, it pays to have some geographic problem-solving techniques in your back pocket. In this paper we illustrate a general approach to finding the closest location to any given US zip code, with a specific, user-accessible example of how to do it, using only Base SAS^®. We also suggest a method for implementing the solution in a production environment, as well as demonstrate how parallel processing can be used to cut down on computing time if there are hardware constraints.

Read the paper (PDF) | Download the data file (ZIP)

Do you write reports that sometimes have missing categories across all class variables? Some programmers write all sorts of additional DATA step code in order to show the zeros for the missing rows or columns. Did you ever wonder whether there is an easier way to accomplish this? PROC MEANS and PROC TABULATE, in conjunction with PROC FORMAT, can handle this situation with a couple of powerful options. With PROC TABULATE, we can use the PRELOADFMT and PRINTMISS options in conjunction with a user-defined format in PROC FORMAT to accomplish this task. With PROC SUMMARY, we can use the COMPLETETYPES option to get all the rows with zeros. This paper uses examples from Census Bureau tabulations to illustrate the use of these procedures and options to preserve missing rows or columns.

Read the paper (PDF) | Watch the recording