topic Questiion on Multiple Linear Regression (proc reg vs. proc glm and model selection) in Statistical Procedures

Questiion on Multiple Linear Regression (proc reg vs. proc glm and model selection)

lousam — Sat, 09 Dec 2017 00:39:19 GMT

Hello,

I am a new SAS user and I need help with clarifying any of the following concepts:

1. When conducting multiple linear regression using mostly continuous predictors, and one categorical predictor that was coded as 0/1 (for instance Sex: 0= male and 1=female), can you use proc reg (since the indicator variable is already created) or do you have to use proc glm with a class statement?

Can I use a code like this:

proc reg data=new;

model BMI = Sex Age Height Weight;

run;

2. Is there a difference between stepwise selection and lasso selection when trying to identify the best fit model?

I appreciate any insight you can offer for any of the questions.

Re: Questiion on Multiple Linear Regression (proc reg vs. proc glm and model selection)

WarrenKuhfeld — Sat, 09 Dec 2017 00:51:09 GMT

Yes you can use PROC REG with a 0/1 variable.

Lasso is one of many variable selection methods. If you are interested in variable selection, see proc glmselect.

Re: Questiion on Multiple Linear Regression (proc reg vs. proc glm and model selection)

KachiM — Sat, 09 Dec 2017 02:59:32 GMT

You may remember:

Since your dependent variable, BMI, is computed using Height and Weight, the inclusion of the latter as independent variables,

will be highly correlated.