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

Re: st: OUTREG2 marginal effect from biprobit???

From   "roy wada" <>
Subject   Re: st: OUTREG2 marginal effect from biprobit???
Date   Mon, 23 Jul 2007 02:32:38 -0700

The number that comes after -mfx- is usually a joint probability. See
-help biprobit postestimation-.

This particular output looks like a univariate probability for varotc=1.
The other variable is not working. -mfx- takes a shortcut by taking
or not taking a derivative something-something of the equation
that works most of the time but not all the time. You can use
-force- option to test it. See this:

You should also try varying your biprobit between SURE and MLE options.
See -help biprobit-. See if -margeff- makes a difference.

Finally, you need to read the manual when performing postestimation.
They exclude the good stuff from the online help for reasons that
would seem obvious but who knows if that's the case.


Hi Roy,
Thank you for your response.
Actually,for your point 1 (I am sorry if I was not clear) I just used
the command: mfx compute and I got the message "warning: predict()
expression  unsuitable for standard-error calculation
> >option nose imposed
> >
> >Marginal effects after biprobit
> >       y  = Pr(vart,varotc=1) (predict)
> >          =  .05270474

and then I got the table with the estimates of marginal effects dy/dx
without the standard errors.

so my question is what is this y=.05270474 that comes before the table?

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index