Re: st: Changing beta coefficients - w/ logistic regression

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

At 05:55 PM 7/21/2005 -0400, you wrote:
> Back in April, someone posted a question about how to change beta
> coefficients.
>
> The answer given (more specifically, the sample code) works perfectly
> for linear regression output. But I can't seem to get it to work when
> the estimated model is a logistic regression.
>
> Any ideas how to make the code work for LOGISTIC regression?
Don't ask me why, but the original foo.ado works fine with ologit and 
gologit2 (both of which, with a dichotomous dv, produce results equivalent 
to logit) - but it bombs with mlogit.  Apparently, some programs rate and 
some do not.  Examples:

. sysuse auto
(1978 Automobile Data)
. quietly ologit foreign mpg weight
. matrix list e(b)

e(b)[1,3]
       foreign:    foreign:       cut1:
           mpg      weight       _cons
y1   -.1685869   -.0039067  -13.708367

. foo
. matrix list e(b)

e(b)[1,3]
       foreign:    foreign:       cut1:
           mpg      weight       _cons
y1         122   -.0039067  -13.708367

. quietly gologit2 foreign mpg weight
. matrix list e(b)

e(b)[1,3]
     Domestic:  Domestic:  Domestic:
          mpg     weight      _cons
y1  -.1685869  -.0039067  13.708367

. foo

. matrix list e(b)

e(b)[1,3]
     Domestic:  Domestic:  Domestic:
          mpg     weight      _cons
y1        122  -.0039067  13.708367

. quietly mlogit foreign mpg weight, b(0)

. matrix list e(b)

e(b)[1,3]
      Foreign:   Foreign:   Foreign:
          mpg     weight      _cons
y1  -.1685869  -.0039067  13.708367

. foo
last estimates not found
r(301);

*
*   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/