# st: Re: how to convert xb into probabilities in probit models

 From "yu chenyu" To statalist@hsphsun2.harvard.edu Subject st: Re: how to convert xb into probabilities in probit models Date Tue, 13 May 2008 14:35:31 +0100

```Dear Maarten,

Thank you very much for your reply. I used Heckman (1981) methods to
correct the initial conditions problem in a dynamic xtprobit model
based on a panel data. There is no built-in command in Stata 10 which
enables me to apply the Heckman's method. So, I used -redprob-
(Stewart,2006) to estimate the model. Could you help me to estimate
probabilities after the model, maybe, the syntax of 'pu0'?

Thank you very much.

Yu

From: Maarten buis <maartenbuis@yahoo.co.uk>
Subject: Re: st: how to convert xb into probabilities in probit models

- --- yu chenyu <ychenyu1@googlemail.com> wrote:
> I want to estimate the probabilities of forming a household in a
> xtprobit model. Because I used a user-written command, both
> '-predict, pu0-' and 'adjust, pr' do not work.

If you have Stata 10 you should use the official -xtprobit- command,
which will allow you to use -predict, pu0-

- -- Maarten

> Hi,
>
> I want to estimate the probabilities of forming a household in a
> xtprobit model. Because I used a user-written command, both '-predict,
> pu0-' and 'adjust, pr' do not work. So, I have to first estimate xb
> coefficients, and then convert xb into probabilities using the normal
> distribution. Could you please tell me how to implement the conversion
> in Stata 10?
>
> Thank you very much.
>
> Best wishes,
> Yu
>
*
*   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/
```