Statalist


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

Re: st: RE: biprobit and restrictions on RHO


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: RE: biprobit and restrictions on RHO
Date   Fri, 20 Jul 2007 22:10:45 +0100 (BST)

You can sometimes make it easier for Stata to converge by rescaling the
explanatory variables in such a way that the estimated parameter are of
approximate equal size (i.e. not a parameter 1000000 and another
parameter .00000000001; you have to multiply the first parameter by
1000000 and divide the second variable by 10000000000). Other things to
look for is categorical variables with a rare category, and
multi-colinearity (high correlation among the explanatory variables).
Usually Stata is well behaved when you have any of these problems, but
if Stata is already strugling any of these things might push it over
the edge.

Hope this helps,
Maarten

--- Nirina F <fstata@gmail.com> wrote:

> Hi Marteen, I tried your example, it worked but once I went to my
> exercise it says not concave and it does not work, even if I tried
> different values.
> 
> also, I don't know what's wrong but one of my regression with
> biprobit
> yesterday worked(i mean I have the output and the logfiles prooving
> that it worked) and today it's giving me not concave.
> grrr!!! (frustrated with the code not anyone:-)
> Thanks in advance for your help.
> Nirina
> 
> 
> On 7/20/07, Maarten Buis <M.Buis@fsw.vu.nl> wrote:
> > --- Nirina F wrote:
> > > I know biprobit can just run but can anyone help if I would like
> to
> > > restrict rho to  a value say 0.3
> >
> > It is possible. You have to know that Stata doesn't maximize with
> > respect to rho, but to the Fisher's Z transformation of rho (also
> > known as the arc-hyperbolic tangent of rho, which is implemented
> > as the -atanh()- function in Stata). So if you want to constrain
> > the rho to be .3, than you have to tell Stata to constrain the
> > appropriate parameter to -atanh(.3)-. The appropriate parameter
> > is the constant in the athrho equation. This is what I have done
> > in the example below (except that I use -.3 instead of .3 as that
> > better fits in this data):
> >
> > *----------- begin example -----------------
> > use http://www.stata-press.com/data/r9/school.dta, clear
> > local athrho = atanh(-.3)
> > constraint 1 [athrho]_cons = `athrho'
> > biprobit priv vote logptax loginc years, constraint(1)
> > *------------ end example ------------------
> > (For more on how to use examples I sent to the Statalist, see
> > http://home.fsw.vu.nl/m.buis/stata/exampleFAQ.html )
> >
> > Hope this helps,
> > Maarten
> >
> > -----------------------------------------
> > Maarten L. Buis
> > Department of Social Research Methodology
> > Vrije Universiteit Amsterdam
> > Boelelaan 1081
> > 1081 HV Amsterdam
> > The Netherlands
> >
> > visiting address:
> > Buitenveldertselaan 3 (Metropolitan), room Z434
> >
> > +31 20 5986715
> >
> > http://home.fsw.vu.nl/m.buis/
> > -----------------------------------------
> >
> >
> >
> >
> > *
> > *   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/
> >
> *
> *   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/
> 


-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

http://home.fsw.vu.nl/m.buis/
-----------------------------------------


      ___________________________________________________________
Yahoo! Answers - Got a question? Someone out there knows the answer. Try it
now.
http://uk.answers.yahoo.com/ 
*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index