Fig. 3
From: Comparing regression modeling strategies for predicting hometime
Distribution of predicted 90-day hometime across the test data set using seven different machine learning models with 15 clinically relevant covariates (A Random forests regression; B Bagged regression trees; C Support vector regression; D Generalized boosting machine (Gaussian distribution, interaction depth = 2); E Generalized boosting machine (Poisson distribution, interaction depth = 15)); F Lasso regression; G Ridge regression