STAT 538 - Statistics for Research II, Regression Modeling

3 Credit Hours

Theory and application of simple and multiple linear regression, including interpretation, diagnostics, model selection, model-building strategies. Causal and predictive approaches are discussed. Logistic regression is introduced for binomial response variables, with focus on estimation, testing, and interpretation. Regression approaches are contrasted with machine learning approaches such as Random Forests. Model evaluation methods are integrated throughout, including techniques for assessing predictive accuracy (RMSE, R², and prediction intervals) and classification performance (accuracy, precision, recall, ROC). Statistical software is required for data analysis, visualization, and implementation of all models. AI tools are presented to assist in data preparation, coding, and summary of results. Students are guided through steps for proper model building through applied case studies.
Registration Restriction(s): Minimum student level - graduate.

Add to Portfolio.

Close Window