HPGENSELECT: parameter estimates per step and lassorho setting

[WeiLSTAT]

I borrow test data in the reference below as an example to explain my questions.

Reference: https://www.mwsug.org/proceedings/2017/AA/MWSUG-2017-AA02.pdf.

Coding was attached to facilitate your check.

Question 1

How should I code to gain parameter estimates in each step of all 20 rather than the selected (last) step alone from a single running of proc hpgenselect? This question is about further calculating other metrics not available in the procedure (such as AUC) per step in data not used to train models.

Snapshot of selection details:

Question 2

Why are parameter estimates, AIC, AICC, BIC, ... not the same between?

• Model 1 Step 2 Lambda 0.8*0.8 = 0.64
• Model 2 Step 1 Lambda 0.64

Why is Chosen Regularization Parameter in Model 2 not equal to 0.64*0.190154?

Model 1

Model 2

Thanks!