Results for training, model selection and validation.

2012-01-12T00:54:51Z (GMT) by Iain H. Moal Paul A. Bates

Results for feature selection, model selection and validation, using the two selection criteria and the four data partitioning schemes. The number of features for the and models is shown (#), alongside their leave-one-out cross-validation correlations and RMSE. The RMSE and correlation of the values used for selecting these models is also shown, as are those when the model is applied to the validation set, along with the significance of correlation.