BSTT 602 : Methods in Biostatistics II

This module describes the logistic function, the popularity of the logistic model, and how the model may be applied in epidemiology. Several important special cases of the logistic model involving dichotomous exposure variable are considered with their corresponding odds ratio expressions. In particular focus, is a simple analysis of a fourfold table, and assessment of multiplicative interaction between two dichotomous variables. Models that account for the potential confounding effects and potential interaction effects of covariates are emphasized. It also describes the general maximum likelihood procedure in estimating the regression parameters. It discusses how to make statistical inferences using maximum likelihood (ML) estimates