Skip to main content

Table 3 Details of the five missing data methods investigated

From: Comparison of imputation methods for handling missing covariate data when fitting a Cox proportional hazards model: a resampling study

Label

Missing data method

CC

Complete case analysis

SI

Single imputation using regression switching imputation with predictive mean matching with only one imputation fitted using the 'pmm' function within the mice library [40]

MI-aregImpute

MI fitting flexible additive imputation models using the 'aregImpute' function in the Hmisc library [21]

MI-MICE

MI using regression switching imputation with linear or logistic regression models as appropriate for each incomplete covariate fitted using the mice library [40]

MI-MICE-PMM

MI using regression switching imputation with predictive mean matching fitted using the 'pmm' function within the mice library [40]

  1. Key: PMM = predictive mean matching; MI = multiple imputation