1 SAS Certificate Applied Statistics and SAS Programming
2 SAS Certificate Applied Statistics and Advanced SAS Programming Brigham Young University Department of Statistics offers an Applied Statistics and Advanced SAS Programming certificate for undergraduate students. The certificate is earned by taking 13 credits in applied statistics and programming from the Department of Statistics. All the courses taken for the SAS certificate also count toward the department core requirements and electives for our bachelor degrees; students will use both SAS and R throughout the course of the program. The certificate credential is not intended as a diploma or a degree. Requirements To earn the certificate, students must complete the following courses with at least a C, but have an overall B grade point average: STAT SAS Certification 1 STAT SAS Certification 2 STAT Statistical Computing 1 STAT Analysis of Variance STAT Introduction to Regression STAT Statistical Computing 2
3 Stat 124: SAS Certification 1 (1 credit) Book: SAS Certification Prep Guide: Base Programming SAS Publishing This course introduces you to the fundamentals of SAS programming. The goal of Statistics 124 is to show you how to use SAS as a programming environment. 'vve focus primarily on the following topics: 1. Accessing Data 2. Creating Data Structures 3. Managing Data 4. Generating Reports 5. Handling Errors The homework provides simple examples so you can develop practical experiences in writing simple SAS programs. Statistics 124 is designed to get participants to take the SAS Base certification exam. Like any programming class this class requires self-discipline and hard work for each student. Be prepared to pass the SAS Certified Base Programmer Exam by demonstrating they can use SAS to: 1. import and export raw data files 2. manipulate and transform data 3. combine SAS data sets 4. create basic detail and summary reports using SAS procedures 5. identify and correct data, syntax and programming logic errors For more information on the SAS Global Certification program and its value in the job market visithttp://support.sas.com/certify/
4 Stat 125: SAS Certification 2 (1 credit) Book: Sharpening Your SAS Skills and SAS Certification Prep Guide: Advanced Programming - SAS Publishing Stat 125 is a one credit hour class taught the second block of Fall and Winter semesters, which was created for the purpose of exposing students to the materials needed to pass the SAS Advanced Programming Certification Exam. **Note that we will not teach you all that is needed to pass the exam. In fact, we will be teaching you very little. Instead, this course should be thought of as an independent study course with resources available to help you in your own study by e.g., outlining what is expected to pass the exam, explaining the types of questions that SAS typically asks on these exams, and answering questions you may have as you work through the SAS Certification Prep Guide alone or with friends. SAS Certified Advanced Programmer Exam A student completing Stat 125: SAS Certification 2 will be prepared to pass the SAS Certified Advanced Programmer Exam by demonstrating they can use SAS to: 1. use advanced DATA step programming statements and efficiency techniques to solve complex problems. 2. write and interpreting SAS SQL code. 3. create and use the SAS MACRO facility. SAS Certified Advanced Programmer certification represents the upper echelon of SAS programmers. Those who earn the advanced programming credential have demonstrated a high level of proficiency in SAS programming expertise and are a much sought after commodity in the global job market. For more information on the SAS Global Certification program visithttp://support.sas.com/certify/ SAS Certified Advanced Programmer Exam A student completing Stat 125: SAS Certification 2 will be prepared to pass the SAS Certified Advanced Programmer Exam by demonstrating they can use SAS to: 1. use advanced DATA step programming statements and efficiency techniques to solve complex problems. 2. write and interpret SAS SQL code. 3. create and use the SAS MACRO facility. SAS Certified Advanced Programmer certification represents the upper echelon of SAS programmers. Those who earn the advanced programming credential have demonstrated a high level of proficiency in SAS programming expertise and are a much sought after commodity in the global job market. For more information on the SAS Global Certification program visithttp://support.sas.com/certify/
5 Stat 224: Statistical Computing 1 (2 credits) Book: Little SAS Book, 3e - Delwiche Statistics 224 is designed for Statistics majors or minors. Like any programming class this class requires selfdiscipline and hard work from each student. This course introduces you to the fundamentals of SAS programming. The goal of Statistics 224 is to get you to use the SAS programming environment. We focus primarily on the following general topics: 1. a lab for Stat DATA step. 3. Macros Programming 4. Report writing (Proc Report) 5. ODS 6. SQL There is some additional SAS Educational Materials that cover the above general topics (basic, progl, prog2, macros, ods, rpros, sqls) that are available for your download and use. 1. Statistics 224 students are expected to show that they can program SAS to: 2. Import and export raw data files 3. Manipulate and transform data using formats, logical and array statements. 4. Combine SAS data sets the Data Step and Proc SQL 5. Write a MACRO program 6. Create detailed summary reports using ODS and SAS procedures, specificly Proc Report 7. Identify and correct data, syntax and programming logic errors
6 Stat 230: Analysis of Variance (3 credits) Book: Introduction to Design and Analysis of Experiments 1e - Cobb Scientific method, statistical thinking, sources of variation, completely randomized design, ANOVA, power and sample size, multiple testing, randomized complete blocks, factorial designs, interactions, statistical software (SAS and R). A student completing Stat 230 will be able to: 1. define and identify experimental unit, response variable, factor(s), and level(s) of a basic experiment. 2. understand the role of randomization and replication in inferring causation. 3. carry out a completely randomized design and construct the ANOVA table by hand and in SAS and R. 4. understand F-tests (what they say, what they do not say). 5. compute the minimum number of replicates in a completely randomized design to achieve a specified level of power. 6. compute confidence intervals for pairwise differences and contrasts of means by hand and in SAS and R. 7. carry out a randomized complete block design and construct the ANOVA table by hand and in SAS and R. 8. carry out a factorial design and construct the ANOVA table in SAS and R. 9. explain the meaning of an interaction. 10. describe the differences between a split-plot and a two-way ANOVA.
7 Stat 330: Introduction to Regression (3 credits) Book: Regression Analysis by Example 4e - Chatterjee Regression, transformations, residuals, indicator variables, variable selection, logistic regression, time series, observational studies, statistical software. A student completing Stat 330 will be able to: 1. explain the difference between ANOVA and regression, and their computational methods. 2. explain the difference in inference between a designed experiment and observational study. 3. fit a regression model in SAS and R. 4. apply appropriate transformations to the response variable to improve agreement with regression assumptions. 5. use residuals and influence diagnostics to assess model fit, agreement with regression assumptions, and identify outliers and influential observations. 6. create sets of indicator variables for categorical explanatory variables. 7. apply step-wise selection to identify a subset regression model that selects the most significant explanatory variables from a large data set. 8. fit a logistic regression model in SAS and R. 9. fit an ARIMA(p, d, q) model and generate forecasts with prediction intervals in SAS and R.
8 Stat 424: Statistical Computing 2 (3 credits) Book: PROC SQL : BEYOND THE BASICS USING SAS Lafler Data Manipulation with R (Use R!) Spector Software for data analysis programming with R Chambers This course fills in part of the gap between study design and data analysis. The course is mostly about data, how to think about your data, how to store your data, and how to work with it preparatory to doing the analysis. This course is also about giving you the tools to use computers to aid you in many different aspects of the statistics field. At the end of Stat 424 the student will be able to: 1. Demonstrate data manipulation and data clearning skills using SQL and the Data Step in SAS, including 2. importing data from external formats such as Excel into SAS, and 3. parsing data from SAS data sets into tables defined on a database (see below). 4. Create and interact with relational databases, including 5. creating a database schema for storing data with well-defined relationships between properly normalized tables, 6. gaining a high level of proficiency with SQL (both from the terminal and within SASs Proc SQL), 7. creating basic summaries by querying a database, and 8. accessing data from a data base for statistical analysis in SAS. 9. Program SAS macros, including programming concepts such as variable declaration, passing parameters by value or reference, Boolean operators, flow control (iterative constructs and conditional statements), and basic understanding of efficient programming skills. 10. Work with large data sets to solve statistical problems using a combination of SAS and SQL skills.
9 SAS Certificate Advanced Statistics
10 SAS Certificate Advanced Statistics Brigham Young University Department of Statistics offers an Advanced Statistics certificate for graduate students. The certificate is earned by taking 15 credits from the Department of Statistics. All the courses taken for the certificate also count toward the department graduate program; students will use both SAS and R throughout the program. The certificate credential is not intended as a diploma or a degree. Requirements To earn the certificate, students must complete the following courses with at least a C, but have an overall B grade point average: STAT Applied Linear Models STAT Modern Regression Methods STAT Statistical Computation STAT Prob Theory & Math Stat 1 STAT Prob Theory & Math Stat 2
11 Stat 535: Applied Linear Models (3 credits) Book: Linear Models in Statistics 2E - Theory of estimation and testing in linear models. Analysis of full-rank model, over-parameterized model, cellmeans model, unequal subclass frequencies, and missing and fused cells. Estimability issues, diagnostics. 1. Students will be able to evaluate and explain when and how to appropriately apply the full-rank and nonfull-rank linear model in the analysis of various kinds of data including balanced, unbalanced, missing, and messy. 2. Students will be able to understand and apply the theory of estimation and testing for Gaussian linear models. 3. Students will be able to generate linear model analyzes in R and SAS with a complete understanding of the input computer code and the output.
12 Stat 536: Modern Regression Methods (3 credits) Book: No Text Weighted least squares, Bayesian linear models, robust regression, nonlinear regression, local regression, generalized additive models, tree-structured regression. This course trains students using statistical methods for modeling a response variable as a function of explanatory variables. Stat 535 (a prerequisite) covered linear models, and this course attempts to cover the complement set. At a minimum you will learn the derivation, computation, and application of the different methods on data. As part of the first-year MS core, this course material will be part of the Comprehensive Written Exam 1. review Linear Regression Models 2. review Weighted Least Squares, Mixed Models 3. Bayesian Linear Regression 4. Measurement Error Models 5. Generalized Linear Models (logistic) 6. Model Assessment and Selection 7. Shrinkage Methods, Bias-Variance Tradeoff, Subset Selection 8. Local Regression (splines, smoothers) 9. Generalized Additive Models (GAM) 10. Tree-based Models, Random Forests 11. Boosting, Bayesian Adaptive Regression Trees 12. P>>n
13 Stat624: Statistical Computation (3 credits) Book: Introduction to Scientific Programming and Simulation Using R Jones C++ for Everyone 2e Horstmann Fundamental numerical methods used by statisticians; programming concepts; efficient use of software available for statisticians; simulation studies. This course trains students to understand computational issues and algorithms most pertinent to statistics. The course can be broken down into roughly four segments: 1. elementary computing concepts including an introduction to Unix, LaTeX, and R; 2. topics in numerical techniques, including root-finding, numerical integration, and optimization (e.g., estimation of parameters in statistical models); 3. methods of data simulation; and 4. an introduction to statistical programming in C. At the conclusion of the course, you should be able to: 1. generate basic LaTeX documents and presentations 2. read in and manipulate data in R 3. generate graphical representations of data in R 4. write simple statistical functions and programs in R and C 5. summarize and write results to file in R and C 6. understand and perform numerical root-finding, integration, and optimization in R and C 7. generate simulated data from complex distributions in R and C
14 Stat641: Prob Theory & Math Stat 1 (3 credits) Book: Statistical Inference 2e - Casella Axioms of probability; combinatorics; random variables, densities and distributions; expectation; independence; joint distributions; conditional probability; inequalities; derived random variables; generating functions; limit theorems; convergence results. Upon successful completion of the course, the student will be able to 1. apply fundamentals of set theory and basic set operations, 2. enumerate the elements of a discrete sample space, 3. solve problems using axioms of probability, conditional probability, independence, and Bayes theorem, 4. describe the properties of the named distributions, 5. manipulate the pdf and cdf of univariate and multivariate discrete and continuous random variables to calculate probabilities and find joint and conditional distributions, 6. find moments and moment generating functions, 7. derive distributions for transformed random variables and order statistics, 8. generate realizations of common probability models using the probability integral transform, 9. use inequalities to create bounds on probabilities and expected values, 10. verify convergence in probability, distribution, and mean square, and 11. prove the Central Limit Theorem (iid and non-identical finite variance versions) and demonstrate it by simulation.
15 Stat 642: Prob Theory & Math Stat 2 (3 credits) Book: Statistical Inference 2e - Casella Introduction to statistical theory; principles of sufficiency and likelihood; point and interval estimation; maximum likelihood; Bayesian inference; hypothesis testing; Neyman-Pearson lemma; likelihood ratio tests; asymptotic results, including delta method; exponential family. On completing this course, the student will have facility with the concepts of statistical theory fundamental to future work in probability and statistics. The student will be able to 1. find sufficient, minimal sufficient, ancillary, and complete statistics, 2. use method of moments, maximum likelihood, and the Bayesian approach to find estimators, 3. evaluate estimators using mean squared error, bias, variance, loss functions, and Monte Carlo methods, 4. apply the Rao-Blackwell Theorem and the Lehmann-Scheff e Theorem to improve existing estimators, 5. derive likelihood ratio tests, Bayesian tests, Wald tests, and score tests, 6. use the Neyman-Pearson Lemma to find UMP tests, 7. evaluate tests with respect to error probabilities and power using analytical, bootstrap, and other Monte Carlo methods, 8. find interval estimators by inverting test statistics, using pivotal quantities, and using the Bayesian approach, 9. evaluate interval estimators with respect to size and coverage probabilities using analytical, bootstrap, and other Monte Carlo methods, 10. evaluate asymptotic properties of estimators with respect to consistency, asymptotic normality, and asymptotic efficiency, 11. describe the asymptotic properites of the MLE, and 12. use the delta method to find asymptotic distributions of transformed random variables.