Fig. 9
From: RRegrs: an R package for computer-aided model selection with multiple regression models
Models’ differences for toxicity training set (data split 2). Differences for R2 and RMSE values among the applied models are presented. The average performance with two-sided confidence limits are plotted as derived by the Student t-test with Bonferroni multiplicity correction. We can observe small differences for LM, LASSO, ENET models, whereas the performance of the best model, SVRM, appears to be close to that of RF and NN models