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

Re: st: cnsreg post estimation- Stata8

From   "Vidya Mahambare" <[email protected]>
To   <[email protected]>
Subject   Re: st: cnsreg post estimation- Stata8
Date   Tue, 28 Jun 2005 15:59:34 +0100

Thanks for both your suggestions.
I am trying to estimate an equation similar to this -

constraint define 1 L.wpi+L2.wpi+L3.wpi=1
cnsreg wpi L.wpi L2.wpi L3.wpi L.ygapm L2.ygapm sswpi L.sswpi, constraint(1)

I couldn't find a way to get robust errors with cnsreg, so I did -

constraint define 1 L.wpi+L2.wpi+L3.wpi=1
arima wpi L.wpi L2.wpi L3.wpi L.ygapm L2.ygapm sswpi L.sswpi, constraint(1)

Both these give the same coefficients, but not other diagonstics such as R2. I
have now looked at FAQ on this that you suggested.

How can use -regress- to implement this constraint?

>>> [email protected] 06/28/05 3:33 PM >>>
At 02:07 PM 6/28/2005 +0100, Vidya Mahambare wrote:
>Could anyone suggest how I can obtain R2/adjusted R2 and other diagnostics 
>-cnsreg- (constrained regression) in Stata8?

Incidentally, I don't know why cnsreg does not provide this 
information.  Many/most/all? models you could estimate with cnsreg could be 
estimated with regular regression, e.g. you could add two variables 
together if you wanted to constrain their effects to be equal.  Often Stata 
has very good reasons for doing things like this, but it doesn't always 
tell you what they are!  Sometimes, too, it takes a conservative strategy, 
e.g. it might be that R^2 would be inappropriate with some constraints, 
ergo it chooses to never give you an R^2.  I encountered something like 
that with the -ml- command, where Stata insisted on reporting a Wald 
chi-square when I used constraints even though I was confident that an LR 
chi-square was appropriate.

Anyway, if cnsreg is making it difficult for you to get things you want, 
you might try using regular regression instead.  You just need to figure 
out how to impose the constraints, e.g. add 2 variables together if you 
want their effects constrained to be equal.

Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX:    (574)288-4373
HOME:   (574)289-5227
EMAIL:  [email protected]
WWW (personal):
WWW (department):

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index