Re: st: Re: LLR test with regress

[Richard Williams <Richard.A.Williams.5@nd.edu>]

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
>
> Kit
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):    http://www.nd.edu/~rwilliam
WWW (department):    http://www.nd.edu/~soc

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/