Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Importing beta coefficients from other softwares


From   Richard Williams <[email protected]>
To   [email protected], [email protected]
Subject   Re: st: Importing beta coefficients from other softwares
Date   Wed, 28 May 2008 13:33:45 -0400

At 12:23 PM 5/28/2008, Maarten buis wrote:
The standard errors alone will not do, you will need the entire
variance covariance matrix. If you have that, you can do something like
this:

*------- begin example ---------------
capture program drop coefin
program define coefin, eclass
    matrix b = 1, 2, 3
    matrix colnames b = foo bar blup
    matrix V = 3, -1, -1 \ ///
               -1, 2, -1 \ ///
               -1, -1, 1
    matrix colnames V = foo bar blup
    matrix rownames V = foo bar blup
    ereturn post b V
    ereturn local cmd "coefin"
end
coefin
nlcom _b[foo]/_b[bar]
This looks good for commands like test and nlcom. If, as was stated originally, other post-estimation commands are desired, e.g. predict, then you may need more or else not be able to do it. In particular, if you want to predict something besides xb or stdp, you would need a predict routine that knew how to do the necessary calculations.

It might help if we knew what Limdep can do that Stata supposedly can't.

I also wonder how hard it would be to do everything with Limdep if for some reason you are forced to use it for the estimation. See

http://www.limdep.com/features/capabilities/post_estimation.php

Excerpts: "All model estimates are 'recoverable.' Coefficients and asymptotic covariance matrices are retained and integrated into the matrix algebra package. Numerical values such as log likelihoods (individual observations and the model criterion function) and sums of squares are recoverable and useable, by name, for example, in testing hypotheses or computing diagnostic statistics... Program tools are provided for manipulating results. One of the most useful is the WALD command which is used to compute nonlinear functions and asymptotic variances for nonlinear functions of model estimates. Programming the derivatives for the delta method is unnecessary."


-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME: (574)289-5227
EMAIL: [email protected]
WWW: http://www.nd.edu/~rwilliam

*
* 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/




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