Regression Analysis

3 credits · Graduate credit

This is an advanced introduction to regression modeling and prediction. Topics covered in this class include linear models for independent observations using least squares estimation (OLS), high dimensional estimation using regularization, regression methods for dependent data including generalized least squares (GLS) and mixed models and generalized linear models (GLM) for single-index modeling of data with a mean/variance relationship. Additional topics may include quantile regression, dimension reduction regression and nonparametric methods for regression and estimation of standard errors including local regression and methods based on reproducing kernel Hilbert spaces (RKHS). The course covers important issues related to data collection, study design, and interpretation of findings, including missing data, non-representative samples, causality, and designed experiments.