Predicting in-hospital-death and mortality percentage using logistic regression

Logistic regression is an appropriate analysis technique for this CinC Challenge problem. Derived variables from provided patient data records are screened for significance by linear stepwise regression. Screened derived variables and corresponding patient outcome data serve respectively as the predictor and response variables for logistic regression analysis. Each of the two CinC Challenge events use separate logistic regression models, and include limited investigation of non-linear effects. Short descriptions of excursions from the logistic regression approach summarize the scope of the effort.