topic How to verify if LINE assumptions are met for a Simple Random Sample using SAS? in Statistical Procedures

How to verify if LINE assumptions are met for a Simple Random Sample using SAS?

coolmac — Wed, 25 Sep 2019 18:31:35 GMT

Trying to find the relationship between AGE (continuous) and GENDER (categorical) variables in my dataset using SAS (v9.4). I used the below code.

proc glm data = myfile;
class GENDER;
model AGE = GENDER / solution;
run;

How to verify if the LINE assumptions of a linear regression model are met or violated in the above case (using SAS)?

L - Linear

I - Independent

N - Normality

E -Equal Variance

Please provide the SAS code to be able to do that.

Reeza — Wed, 25 Sep 2019 18:44:16 GMT

With this case, your question simplifies to the assumptions of an ANOVA/T-test so you're mostly ok as long as your variances are similar.

You can use PROC NPAR1WAY to assess distributions.

To test for equal variances see this example:
https://documentation.sas.com/?docsetId=statug&docsetVersion=15.1&docsetTarget=statug_glm_examples10.htm&locale=en

Note the PLOT=Diagnostics which will provide some diagnostic plots.