Re: st: How to get marginal effect of quadratic term in tobit model in stata 11.0

[lutunnahar begum <pikitardif@gmail.com>]

Hi Maarten
Thanks...that's really a good way to see the non-linear effect. I will
definitely try that.

I tried the following command:

gen tb2=tb^2
tobit ch tb tb2 .........., ll(0) robust
margins, dydx(*) predict (ystar (0,.)) post

I also tried -dtobit- with -version 8.2- and got the 4 MEs  for squared term.

I understand that the true ME should not consider the squared effect
separately as there is only one variable. But papers using tobit model
report the marginal effect of squared term. I am confused whether I
should use the above mentioned command to get this.

But what will be the interpretation?

Thanks for your help.

Regards,
Nahar





On Mon, Jul 18, 2011 at 9:18 PM, Maarten Buis <maartenlbuis@gmail.com> wrote:
> On Mon, Jul 18, 2011 at 1:07 PM, lutunnahar begum wrote:
>> I am using stata 11.0. I am interested in non-linear effect of some of
>> the regressors in tobit model....but for tobit, we don't get any
>> marginal effect for quadratic term.
>
> And that is exactly as it should be. Conceptually, adding a squared
> term of a variable x only means that x has a non-linear effect. So
> there is only one variable x. The fact that the the variable x is
> entered in the model as two "variables", x and x^2, does not change
> the fact that from a substantive point of view there is only one
> variable. Maybe you find the following helpful:
>
> <http://www.maartenbuis.nl/publications/inter_quadr.html>
>
> Hope this helps,
> Maarten
>
> --------------------------
> 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/