I have been working on LASSO in SAS lately, and I'm still trying to figure out how to work with the options, but my main question for which I have not been able to find an answer on the internet so far is: how do you interpret the SAS output?

For example (using SAS website examples):

Lasso summary

I know that in this one, you can see what has been entered and removed, but I'm not sure about CP.

Lasso 2

Here I know it tells me which variables have been retained, but I'm not sure if I need to report anything from this or what it means.

enter image description here

Do these numbers represent the effect each variable has on the outcome? What numbers tell me if this is a good model or if the variables chosen with LASSO are actually good predictors? Thanks.


Cp is the method you chose for selecting the best fit. I think Cp is the default. The * next to one Cp indicates the best fit.

The parameter estimates table is parameter estimates, just like in PROC REG or GLM. However, it is often better to use GLMSELECT to choose one or several models for further exploration and then do more work in PROC GLM or REG, which have many more options for graphs and so on. You can use the best model chosen by GLMSELECT and then evaluate it as you usually would.

GLMSELECT is intended for picking a good model (or several).

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.