Specifying the functional form of a covariate is a fundamental part of developing a regression model. The choice to include a variable as continuous, categorical, or as a spline can be determined by model fit. This paper offers an efficient and user-friendly SAS^® macro (%SPECI) to help analysts determine how best to specify the appropriate functional form of a covariate in a linear, logistic, and survival analysis model. For each model, our macro provides a graphical and statistical single-page comparison report of the covariate as a continuous, categorical, and restricted cubic spline variable so that users can easily compare and contrast results. The report includes the residual plot and distribution of the covariate. You can also include other covariates in the model for multivariable adjustment. The output displays the likelihood ratio statistic, the Akaike Information Criterion (AIC), as well as other model-specific statistics. The %SPECI macro is demonstrated using an example data set. The macro includes the PROC REG, PROC LOGISTIC, PROC PHREG, PROC REPORT, PROC SGPLOT, and more procedures in SAS^® 9.4.

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

Visual+D2:D18ization is a critical part of turning data into knowledge. A customized graph is essential to make data visualization meaningful, powerful, and interpretable. Furthermore, customizing grouped data into a desired layout with specific requirements such as clusters, colors, symbols, and patterns for each group can be challenging. This paper provides a start-from-scratch, step-by-step solution to create a customized graph for grouped data using SAS^® Graph Template Language (GTL). By analyzing the data and target graph with the available tools and options that GTL provided, this paper demonstrates GTL is a powerful and flexible tool to create a customized, complex graph.

Data is generated every second. The term big data refers to the volume, variety, and velocity of data that is being produced. Now woven into every sector, its size and complexity has left organizations faced with difficulties in being able to create, manipulate, and manage big data. This research identifies and reviews a range of big data techniques within SAS^®, highlighting the fundamental opportunities that SAS provides for overcoming a variety of business challenges. Insurance is a data-dependent industry. This research focuses on understanding what SAS can offer to insurance companies and how it could interact with existing customer databases and online, user-generated content. A range of data sources have been identified for this purpose. The research demonstrates how models can be built based on existing relationships found in past data and then used to identify prospective customers. Principal component analysis, cluster analysis, and neural networks are all considered. You will learn how these techniques can be used to help capture valuable insight, create firm relationships, and support customer feedback. Whether it is prescriptive, predictive, descriptive, or diagnostic analytics, harnessing big data can add background and depth, providing insurance companies with a more complete story. You will see that you can reduce the complexity and dimensionality of data, provide actionable intelligence, and essentially make more informed business decisions.

Customer churn is an important area of concern that affects not just the growth of your company, but also the profit. Conventional survival analysis can provide a customer's likelihood to churn in the near term, but it does not take into account the lifetime value of the higher-risk churn customers you are trying to retain. Not all customers are equally important to your company. Recency, frequency, and monetary (RFM) analysis can help companies identify customers that are most important and most likely to respond to a retention offer. In this paper, we use the IML and PHREG procedures to combine the RFM analysis and survival analysis in order to determine the optimal number of higher-risk and higher-value customers to retain.

This presentation discusses the options for including continuous covariates in regression models. In his book, 'Clinical Prediction Models,' Ewout Steyerberg presents a hierarchy of procedures for continuous predictors, starting with dichotomizing the variable and moving to modeling the variable using restricted cubic splines or using a fractional polynomial model. This presentation discusses all of the choices, with a focus on the last two. Restricted cubic splines express the relationship between the continuous covariate and the outcome using a set of cubic polynomials, which are constrained to meet at pre-specified points, called knots. Between the knots, each curve can take on the shape that best describes the data. A fractional polynomial model is another flexible method for modeling a relationship that is possibly nonlinear. In this model, polynomials with noninteger and negative powers are considered, along with the more conventional square and cubic polynomials, and the small subset of powers that best fits the data is selected. The presentation describes and illustrates these methods at an introductory level intended to be useful to anyone who is familiar with regression analyses.

Randomized control trials have long been considered the gold standard for establishing causal treatment effects. Can causal effects be reasonably estimated from observational data too? In observational studies, you observe treatment T and outcome Y without controlling confounding variables that might explain the observed associations between T and Y. Estimating the causal effect of treatment T therefore requires adjustments that remove the effects of the confounding variables. The new CAUSALTRT (causal-treat) procedure in SAS/STAT^® 14.2 enables you to estimate the causal effect of a treatment decision by modeling either the treatment assignment T or the outcome Y, or both. Specifically, modeling the treatment leads to the inverse probability weighting methods, and modeling the outcome leads to the regression methods. Combined modeling of the treatment and outcome leads to doubly robust methods that can provide unbiased estimates for the treatment effect even if one of the models is misspecified. This paper reviews the statistical methods that are implemented in the CAUSALTRT procedure and includes examples of how you can use this procedure to estimate causal effects from observational data. This paper also illustrates some other important features of the CAUSALTRT procedure, including bootstrap resampling, covariate balance diagnostics, and statistical graphics.

Student growth percentile (SGP) is one of the most widely used score metrics for measuring a student's academic growth. Using longitudinal data, SGP describes a student's growth as the relative standing among students who had a similar level of academic achievement in previous years. Although several models for SGP estimation have been introduced, and some models have been implemented with R, no studies have yet described using SAS^®. As a result, this research describes various types of SGP models and demonstrates how practitioners can use SAS procedures to fit these models. Specifically, this study covers three types of statistical models for SGP: 1) quantile regression-based model 2) conditional cumulative density function-based model 3) multidimensional item response theory-based model. Each of the three models partly uses procedures in SAS, such as PROC QUANTREG, PROC LOGISTIC, PROC TRANSREG, PROC IRT, or PROC MCMC, for its computation. The program code is illustrated using a simulated longitudinal data set over two consecutive years, which is generated by SAS/IML^®. In addition, the interpretation of the estimation results and the advantages and disadvantages of implementing these three approaches in SAS are discussed.

SAS/STAT^® software has several procedures that estimate parameters from generalized linear models designed for both continuous and discrete response data (including proportions and counts). Procedures such as LOGISTIC, GENMOD, GLIMMIX, and FMM, among others, offer a flexible range of analysis options to work with data from a variety of distributions and also with correlated or clustered data. SAS^® procedures can also model zero-inflated and truncated distributions. This paper demonstrates how statements from PROC NLMIXED can be written to match the output results from these procedures, including the LS-means. Situations arise where the flexible programming statements of PROC NLMIXED are needed for other situations such as zero-inflated or hurdle models, truncated counts, or proportions (including legitimate zeros) that have random effects, and also for probability distributions not available elsewhere. A useful application of these coding techniques is that programming statements from NLMIXED can often be directly transferred into PROC MCMC with little or no modification to perform analyses from a Bayesian perspective with these various types of complex models.

The analysis of longitudinal data requires a model that correctly accounts for both the inherent correlation amongst the responses as a result of the repeated measurements, as well as the feedback between the responses and predictors at different time points. Lalonde, Wilson, and Yin (2013) developed an approach based on generalized method of moments (GMM) for identifying and using valid moment conditions to account for time-dependent covariates in longitudinal data with binary outcomes. However, the model developed using this approach does not provide information about the specific relationships that exist across time points. We present a SAS^® macro that extends the work of Lalonde, Wilson, and Yin by using valid moment conditions to estimate and evaluate the relationships between the response and predictors at different time periods. The performance of this method is compared to previously established results.

In a randomized study, subjects are randomly assigned to either a treated group or a control group. Random assignment ensures that the distribution of the covariates is the same in both groups and that the treatment effect can be estimated by directly comparing the outcomes for the subjects in the two groups. In contrast, subjects in an observational study are not randomly assigned. In order to establish causal interpretations of the treatment effects in observational studies, special statistical approaches that adjust for the covariate confounding are required to obtain unbiased estimation of causal treatment effects. One strategy for correctly estimating the treatment effect is based on the propensity score, which is the conditional probability of the treatment assignment given the observed covariates. Prior to the analysis, you use propensity scores to adjust the data by weighting observations, stratifying subjects that have similar propensity scores, or matching treated subjects to control subjects. This paper reviews propensity score methods for causal inference and introduces the PSMATCH procedure, which is new in SAS/STAT^® 14.2. The procedure provides methods of weighting, stratification, and matching. Matching methods include greedy matching, matching with replacement, and optimal matching. The procedure assesses covariate balance by comparing distributions between the adjusted treated and control groups.

There is an industry-wide push toward making workflows seamless and reproducible. Incorporating reproducibility into the workflow has many benefits; among them are increased transparency, time savings, and accuracy. We walk through how to seamlessly integrate SAS^®, LaTeX, and R into a single reproducible document. We also discuss best practices for general principles such as literate programming and version control.

Survival analysis differs from other types of statistical analysis, including graphical summaries and regression modeling procedures, because data is almost always censored. The purpose of this project is to apply survival analysis techniques in SAS^® to practical survival data, aiming to understand the effects of gender and age on lung cancer patient survival at different cancer sites. Results show that both gender and age are significant variables in predicting lung cancer patient survival using the Cox proportional hazards model. Females have better survival than males when other variables in the model are fixed (p-value 0.0254). Moreover, the hazard of patients who are over 65 is 1.385 times that of patients who are under 65 (p-value 0.0145).

Socioeconomic status (SES) is a major contributor to health disparities in the United States. Research suggests that those with a low SES versus a high SES are more likely to have lower life expectancy; participate in unhealthy behaviors such as smoking and alcohol consumption; experience higher rates of depression, childhood obesity, and ADHD; and experience problems accessing appropriate health care. Interpreting SES can be difficult due to the complexity of data, multiple data sources, and the large number of socioeconomic and demographic measures available. When SES is expanded to include additional social determinants of health (SDOH) such as language barriers and transportation barriers to care; access to employment and affordable housing; adequate nutrition, family support and social cohesion; health literacy; crime and violence; quality of housing; and other environmental conditions, the ability to measure and interpret the concept becomes even more difficult. This paper presents an approach to measuring SES and SDOH using publicly available data. Various statistical modeling techniques are used to define state-specific composite SES scores at local areas-ZIP Code and Census Tract. Once developed, the SES/SDOH models are applied to health care claims data to evaluate the relationship between health services utilization, cost, and social factors. The analysis includes a discussion of the potential impact of social factors on population risk adjustment.

Chemical incidents involving irritant chemicals such as chlorine pose a significant threat to life and require rapid assessment. Data from the Validating Triage for Chemical Mass Casualty Incidents A First Step R01 grant was used to determine the most predictive signs and symptoms (S/S) for a chlorine mass casualty incident. SAS^® 9.4 was used to estimate sensitivity, specificity, positive and negative predictive values, and other statistics of irritant gas syndrome agent S/S for two exiting systems designed to assist emergency responders in hazardous material incidents (Wireless Information System for Emergency Responders (WISER) and CHEMM Intelligent Syndrome Tool (CHEMM-IST)). The results for WISER showed the sensitivity was .72 to 1.0; specificity .25 to .47; and the positive predictive value and negative predictive value were .04 to .87 and .33 to 1.0, respectively. The results for CHEMM-IST showed the sensitivity was .84 to .97; specificity .29 to .45; and the positive predictive value and negative predictive value were .18 to .42 and .86 to .97, respectively.

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

Bivariate Cox proportional models are used when we test the association between a single covariate and the outcome. The test repeats for each covariate of interest. SAS^® uses the last category as the default reference. This raises problems when we want to keep using 0 as our reference for each covariate. The reference group can be changed in the CLASS statement. But, if a format is associated with a covariate, we have to use the corresponding format instead of raw numeric data. This problem becomes even worse when we have to repeat the test and manually enter the reference every single time. This presentation demonstrates one way of fixing the problem using the MACRO function and SYMPUT function.

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