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

st: Re: Question about mvtobit

From   mikkel barslund <>
Subject   st: Re: Question about mvtobit
Date   Wed, 10 Oct 2007 17:37:47 +0200

On 10/8/07, <> wrote:
> I am forwarding your e-mail to the author of -mvtobit-. Statalist members can read in cc.
> There seems to be a bug, because I obtained exactly the same error message by applying the following code to my dataset: mvtobit (brev = inv1 logstud) (brev1 = inv1 logstud)
> However, as you can see, the code (and to tell the truth, my dataset as well) does not contain any variable named y1!!!
> Nicola

There was an embarrassing bug in the first version of -mvtobit- which
has been corrected in an update.
I responded to private correspondence by Hung-Hao Chang but failed to
inform statalist.
Please, let me know if you observe any other problems.
Thanks for the feedback.

> >From: "Hung-Hao Chang" <>
> >
> >Dear Nicola:
> >  I saw your response to the statalist about SUR & MVTOBIT model. I have downloaded the command and had difficulty to use it. I am wondering if you have better experience to share. Following is my code:
> >
> >mvtobit (y1= x1 x2 x3 x4) (y2=x1 x2 x3 x4) (y3=x1 x2 x3 x4)
> >
> >The error massage is "Fitting univariate Tobit models to get starting values.  y1 not found. r(111);"
> >
> >However, I can estimate each equation in tobit. So this should not a problem of variable name.
> >
> >I appreciate if you can give me a hand. Thanks.
> >
> >Hung-Hao Chang
> >Department of Agricultural Economics
> >National Taiwan University

*   For searches and help try:

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