Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

# st: 回复： st: set initial values for ml model

 From 健 张 To "statalist@hsphsun2.harvard.edu" Subject st: 回复： st: set initial values for ml model Date Tue, 20 Sep 2011 04:13:59 +0800 (CST)

```Hi, Maarten
Thanks again for your kindly help! For the error report, I just search the previous post of other people on the statalist. I find that one post has exactly the same question as mine and you answered her. Here is the post and your answer  http://www.stata.com/statalist/archive/2006-12/msg00322.html

In your response, you mentioned: "I've posted three times an example that shows how to create the correct matrix and column and equation names for initial values." Where can I find that example you mentioned?

Best,
Jian

----- 原邮件 -----

That depends on way too many things. You just need to find why the
matrix you think should contain the starting values, does in fact not
accordingly.

-- Maarten

2011/9/18 健 张 <victerzj2@yahoo.com.cn>:
> Hi, all:
> I am trying to use ml to estimate a bivariate probit model so that I can set initial values myself. First, I use biprobit to estimate the two equations and then I got all coefficients stored in e(b). Then I use ml to write my own likelihood evaluator. I use the following code to set initial values:
> matrix b0=e(b)
> ml init b0
> However, stata always reports that " initial vector: extra parameter treated:connect1 found specify skip option if necessary". Treated is the dependent variable of my first equation and connect1 is a variable listed in both equations.
>
> In addition, anyone knows how to find a good initial value? Since I have around 20 independent variables in total in two equations, it is hard for me to fish the good initial value one by one.
>
>
>
> Best,
> Jian
>
> *
> *   For searches and help try:
> *  http://www.stata.com/help.cgi?search
> *  http://www.stata.com/support/statalist/faq
> *  http://www.ats.ucla.edu/stat/stata/
>

--
--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://www.maartenbuis.nl
--------------------------

*
*   For searches and help try:
*  http://www.stata.com/help.cgi?search
*  http://www.stata.com/support/statalist/faq
*  http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
```