OA3103 Data Analysis

Techniques for analyzing, summarizing and comparing sets of real data with several variables focusing on linear models. Computations are done using statistical software. Model building and verification, graphical methods of exploration. Analysis of variance including multiple comparisons, with coverage of interaction and simple design cases.  Least squares and robust regression, diagnostics, selection of transformations, and variable selection techniques. PREREQUISITE: OA3102.

Prerequisite

OA3102

Lecture Hours

4

Lab Hours

1

Course Learning Outcomes

Increase proficiency with the R language
• Understand and apply ANOVA methodology
• Know how to address multiple testing issues
• Understand the concept of interaction
• Understand blocking and how to handle it
• Use vector-matrix notation and understand matrix concepts in probability and statistics applications
• Understand the simple and multiple regression models and their properties
• Know how to apply classical inference for multiple regression
• Know how to use diagnostics for multiple linear regression
• Understand what outliers are, and how to detect and handle them
• Know how to apply transformations to predictor and outcome variables
• Know how to apply model selection methods including stepwise, best subsets, and other variable-selection techniques
• Know how to include categorical variables in regression