Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Re: LLR test with regress

From   Richard Williams <>
Subject   Re: st: Re: LLR test with regress
Date   Sun, 28 Dec 2003 21:59:57 -0500

At 08:05 PM 12/28/2003 -0500, Christopher F Baum wrote:
webuse auto
constr def 1 price=weight
cnsreg mpg price weight, c(1)
est store constr
reg mpg price weight
lrtest constr, force

That looks like a good alternative when you are using linear regression. There is nothing like a -cnslogit- command, right? I wonder why not? Rather than a separate -cnsreg- command, it seems like it would be optimal to have a -constraints- parameter that could be used with all estimation commands.

How about the general issue of using -lrtest- rather than -test- when doing linear regression? Is it considered appropriate? Is -test- considered more optimal? My experience has been that -lrtest- and -test- tend to give similar, but not identical, results, when used with regards to the -regress- command. As was discussed in a thread a few days ago, if you are doing something like -logit-, -lrtest- is generally better than -test-, but I don't know if this is also true (or the opposite is true) for linear regression. For my own part, I've always used the sort of F tests provided by -test- when testing hypotheses involving linear regression. (Until I started using Stata, I didn't even realize there might be an alternative.)

Richard Williams, Associate Professor
OFFICE: (574)631-6668, (574)631-6463
FAX: (574)288-4373
HOME: (574)289-5227
EMAIL: Richard.A.Williams.5@ND.Edu
WWW (personal):
WWW (department):

* For searches and help try:

© Copyright 1996–2023 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index