Fig. 1
From: Censcyt: censored covariates in differential abundance analysis in cytometry
Single cluster simulation results for a sample size of 100 for censoring rates of 30%, 50% and 70%. Shown are four measures calculated from 100 simulation repetitions: raw bias (RB), coverage rate (CR), confidence interval (CI) width and root mean squared error (RMSE). cc: complete case analysis, km: Kaplan–Meier imputation, kme: Kaplan–Meier imputation with an exponential tail, mrl: mean residual life imputation (conditional multiple imputation), pmm: predictive mean matching (treating censored values as missing), rs: risk set imputation. Other parameter values are: true regression coefficient \(\beta _1=-\,1e{-}4\), number of multiple imputations \(=\,50\) and the variance of the random effect \(=\,1\)