Skip to main content

Table 3 The proportion at which a given variable is selected in the training set and is significant in the validation set after randomly splitting the data 1000 times

From: Determining relative importance of variables in developing and validating predictive models

Variables Forward
selection
Backward
selection
Stepwise
selection
  sv sv sv
Age 0.979 0.956 0.954
Baseline WBC 0.873 0.803 0.789
T-cell immunophenotype 0.394 0.464 0.464
Mediastinal mass 0.248 0.270 0.273
Splenomegay 0.192 0.118 0.136
Hepatomegaly 0.016 0.000 0.000
Sex 0.000 0.000 0.000