Multiple imputation of partially observed covariates in discrete-time survival analysis

Discrete-time survival analysis (DTSA) models are a popular way of modeling events in the social sciences. However, discrete-time data is challenged by missing one or more covariates. Negative consequences covariate include efficiency losses and possible bias. A approach to circumventing these multiple imputation (MI). In MI, it crucial outcome information models. As there little guidance on how incorporate observed into model covariates DTSA, we explore different existing approaches using fully conditional specification (FCS) MI substantive-model compatible (SMC)-FCS MI. We extend SMC-FCS for DTSA provide an implementation smcfcs R package. compare Monte Carlo simulations demonstrate good performance new compared approaches.