Fig. 5.
Glioma data. The effect of the choice of the number of variables without zero weights in the penalty on the mean difference in canonical correlation between the training and the validation set (black) and the training and the permutation sets (gray) for three succeeding canonical variate pairs. Each line represents the effect of the number of selected variables in that set, while the number of selected variables in the other set stays fixed. In the right column, the effect of the number of SNP variables and in the left the effect of the number of gene expression variables.

Glioma data. The effect of the choice of the number of variables without zero weights in the penalty on the mean difference in canonical correlation between the training and the validation set (black) and the training and the permutation sets (gray) for three succeeding canonical variate pairs. Each line represents the effect of the number of selected variables in that set, while the number of selected variables in the other set stays fixed. In the right column, the effect of the number of SNP variables and in the left the effect of the number of gene expression variables.

Close
This Feature Is Available To Subscribers Only

Sign In or Create an Account

Close

This PDF is available to Subscribers Only

View Article Abstract & Purchase Options

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Close