Northwestern University Feinberg School of Medicine

Department of Preventive Medicine


The software modules listed below were written by members of the Biostatistics Division within the Department of Preventive Medicine of Northwestern University Feinberg School of Medicine.   A brief description is provided for each module.

This software is free, however it is provided without any warranty or implied warranty of fitness, correctness, or use for a particular purpose.   Use at your own risk. 

SAS Macros

ADJSURV_CHR - A SAS macro for calculating cumulative hazard ratios and direct-adjusted survival probabilities for K treatment groups based on a stratified Cox regression model. Based on papers by Zhang et al (2007) and Wei and Schaubel (2008), this macro allows one to compare adjusted survival between K treatment groups when the assumption of proportional hazards is violated.

MIDAS - a SAS macro for multiple imputation using distance aided selection of donors. Implements an iterative predictive mean matching hot-deck for imputing missing data

R functions

Binary_imp - function determining order of assignment of binary variables from underlying normally distributed data based on quantiles for entries where two or more values had to be imputed.

CombineNestedImputations - An R function for combining inferences based on nested multiple imputations. This R function combines inferences based on nested multiply imputed data sets and calculates rates of missing information.

Mixed_imp - Function for assigning binary values based on quantiles of underlying normally distributed values.

Modified Lurie-Goldberg Code - This code allows for the multiple imputation of multivariate continuous data using a semi-parametric approach, allowing one to relax distributional assumptions of the data, as described in Helenowski and Demirtas (2013), forthcoming in the Journal of Biopharmaceutical Statistics.