Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Brant test and goodness of fit(after -oprobit-) with complex survey data


From   Richard Williams <richardwilliams.ndu@gmail.com>
To   statalist@hsphsun2.harvard.edu, statalist@hsphsun2.harvard.edu
Subject   Re: st: Brant test and goodness of fit(after -oprobit-) with complex survey data
Date   Wed, 06 Jun 2012 11:42:44 -0500

At 03:10 AM 6/6/2012, Maarten Buis wrote:
On Wed, Jun 6, 2012 at 9:13 AM, Sol <ykim.berlin@gmail.com> wrote:
> 1. From what I've read, the Brant test of parallel lines assumption can be
> tested in Stata with -omodel probit-. The problem is that this command
> doesn't work with the -svy- syntax(I get the message "omodel is not
> supported by svy with vce(linearized)"). Is there another way the test can
> be implemented in Stata?

If I remember correctly you can use -gologit2- to get a test for
parallel lines and -gologit2- allows svy. You can install -gologit2-
by typing in Stata -ssc install gologit2- and you can read more about
it here: <http://www.nd.edu/~rwilliam/gologit2/>

Hope this helps,
Maarten

Correct. You can test individual variables using the -autofit- option. For a global test, similar to Brant, do something like

webuse nhanes2f, clear
gologit2 health female black age height weight, svy
test [#1 = #2 = #3 = #4]

If you weren't using svy, you could do a LR chi-square test, e.g.

gologit2 health female black age height weight, pl sto(ologit)
gologit2 health female black age height weight, npl sto(gologit)
lrtest ologit gologit, stats

Maarten and I did some simulations a few years ago where we compared brant, omodel, and gologit2. All performed about the same, at least in our simulations. brant appeared to be the best. omodel tended to not reject the null often enough, brant was about right, gologit2 rejected a little too often. gologit2 is the only approach I know of when data are svyset. gologit2 also supports several different link functions.

If you have a data set that uses some of the svyset options introduced after Version 9 of stata, I have a beta version of gologit2 that can handle them. Write me off-list for details.


-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW:    http://www.nd.edu/~rwilliam

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index