Abstract
This paper reviews concepts of multivariate statistical modeling via the logistic regression model, which has become very popular for modeling the relationship between a positive clinical outcome and a variety of predictor variables. The process is illustrated using a composite of data from three large prostate specific antigen based screening studies of prostate cancer.

This publication has 0 references indexed in Scilit: