Date:         Mon, 26 Aug 2002 10:35:29 -0400
Reply-To:     Doc Muhlbaier <lawrence.muhlbaier@DUKE.EDU>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Doc Muhlbaier <lawrence.muhlbaier@DUKE.EDU>
Organization: Duke University, Durham, NC, USA
Subject:      Re: lilliefors test

Doug,

The Lillefors test is a special case of the Kolmogorov-Smirnov test and
available in the NORMAL option of PROC UNIVARIATE.

In general, it is not a very powerful test, so if it does not reject, that
is not very strong evidence that the data are normally distributed.

Doc

"Doug" <queanbeyan@hotmail.com> wrote in message
news:3d6a19ed@dnews.tpgi.com.au...
> Hi
>
> Could someone please tell me how I can carry out a lilliefors test in SAS.
> I need to check some residuals from a proc glm for normality.  I know that
I
> can do a normal plot in univariate - but I can't find any information
about
> lilliefors in SAS Stat.
>
> Is the test simply the same as the kolmogorov-smirnov for normal
> distribution, and if so is it simply a matter of rejecting the hypothesis
of
> normality if the p-value is less than alpha?
>
> Doug
>
>