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

Re: st: biprobit: set correlation of residuals to zero

From   Maarten buis <[email protected]>
To   [email protected]
Subject   Re: st: biprobit: set correlation of residuals to zero
Date   Mon, 1 Sep 2008 17:11:24 +0100 (BST)

--- "Karfakis, Panagiotis (ESAF)" <[email protected]> wrote:
> is there any way I could set a constraint in biprobit so that the
> correlation of the residuals is zero?

In that case you can just estimate separate -probit-s. Alternatively
you can use -constraint-. In that case you need to know that 
-biprobit- does not maximizes the likelihood with respect to the
correlation coefficient, but with respect to the Fischer's Z
transformed correlation. However, the Fischer's Z transformed 0 is
still 0, so that is no problem in this case. Below is an example of how
to do all this in Stata.

*------------------ begin example ------------------------
webuse school
biprobit private vote logptax loginc years
constraint 1 [athrho]_cons=0
biprobit private vote logptax loginc years, constraint(1)
probit private logptax loginc years
probit vote logptax loginc years
*------------------- end example -------------------------
(For more on how to use examples I sent to the Statalist, see )

Hope this helps,

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

*   For searches and help try:

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