Fig. 1From: A flexible approach for variable selection in large-scale healthcare database studies with missing covariate and outcome dataThe mean cross-validated AUC, averaged across 250 data replications, for each of three methods: RR-BART, BI-BART and BI-XGB. The mean AUC for bootstrap imputation based methods BI-BART and BI-XGB varies by the threshold value of π. missForest was used for imputation. The sample size n=1000. The proportion of missingness is 40% in the outcome Y and is 60% overallBack to article page