Re: multiple regression assumptions SAS
- From: "Paige Miller" <paige.miller@xxxxxxxxx>
- Date: 20 Mar 2007 12:39:50 -0700
On Mar 18, 11:27 pm, nko...@xxxxxxxxx wrote:
Those of you familiar with SAS, how would I test for the multiple
regression assumptions?
To test the assumption in multiple regression is that the errors are
normally distributed, you would output the residuals from PROC REG
into a SAS data set using the OUTPUT statement in PROC REG.
Then you could test normality any one of a number of ways. You could
test it graphically by creating histograms in PROC GCHART, PROC
CAPABILITY or PROC UNIVARIATE. These last two procedures will also do
a formal test to see if the residuals are normally distributed.
--
Paige Miller
paige\dot\miller \at\ kodak\dot\com
.
- References:
- multiple regression assumptions SAS
- From: nkomli
- multiple regression assumptions SAS
- Prev by Date: Re: WinBugs Conditional Formulation
- Next by Date: scale of logit model
- Previous by thread: multiple regression assumptions SAS
- Next by thread: Sum of two independent uniform random variables
- Index(es):
Relevant Pages
|