04-03-2017 01:41 AM
I am trying to plot LASSO solution pass such as Figure 3 of "Funda Gunes, SAS Institute Inc. Penalized Regression Method for Linear Models in SAS/STATS".
"plots=all" option of glmselect shows the graph of which X axis shows step No., Y axis shows ASE (Aberage Squared Errors)
but I want to plot the graph of which X axis shows the normalized tuning value (L1 norm of the regression coefficients divided by the L1 norm of the ordinary least squares splution), Y azis shows ASE.
How can I plot it ?