Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.

# Re: st: Testing whether the average partial effects are statistically different across groups?

 From Tirthankar Chakravarty To statalist@hsphsun2.harvard.edu Subject Re: st: Testing whether the average partial effects are statistically different across groups? Date Wed, 31 Aug 2011 05:06:46 -0700

```The above example extends simply by adding -xt- to -probit-:

/**************************************/
webuse union, clear
xtprobit union age grade not_smsa ///
i.south##c.year
margins, over(south) dydx(c.year) ///
at(south=(0 1)) coeflegend post
test _b[1bn._at#0bn.south]= _b[2._at#1.south]
/**************************************/

T

On Wed, Aug 31, 2011 at 4:13 AM, natasha agarwal
> Thank you very much for your kind assistance.
>
> Can this also be performed after xtprobit, re?
>
> Thanks
> Natasha
>
> On Wed, Aug 31, 2011 at 12:05 PM, Tirthankar Chakravarty
> <tirthankar.chakravarty@gmail.com> wrote:
>> Actually, I think what you want is something slightly different, in
>> that you only want to compare the AME for group1 for the group1
>> subsample, with the AME for group2 for the group2 subsample. This can
>> be had by decomposing the marginal effects as discussed in the manual
>> entry [R] on -margins-, pg. 969-972. Which makes more sense depends on
>> the nature of the "e" variable.
>>
>> /**************************************/
>> webuse union, clear
>> probit union age grade not_smsa ///
>>        i.south##c.year
>> margins, over(south) dydx(c.year) ///
>>  at(south=(0 1)) coeflegend post
>> test _b[1bn._at#0bn.south]= _b[2._at#1.south]
>> /**************************************/
>>
>> T
>>
>>
>> On Wed, Aug 31, 2011 at 3:49 AM, Tirthankar Chakravarty
>> <tirthankar.chakravarty@gmail.com> wrote:
>>> Create an indicator variable for your condition (i.e., a variable
>>> which is 1 if e>=0.5, 0 else), interact it fully with the variable of
>>> interest ("fdi"), compute -margins, over(indicator) dydx(fdi)-, that
>>> is the average marginal effects of "fdi" over the values of the
>>> indicator term, and test the values produced for equality, as in the
>>> following example:
>>>
>>> /**************************************/
>>> webuse union, clear
>>> probit union age grade not_smsa ///
>>>        i.south##c.year
>>> margins, over(south) dydx(c.year)
>>> test _b[0bn.south]=_b[1.south]
>>> /**************************************/
>>>
>>> T
>>>
>>> On Wed, Aug 31, 2011 at 3:35 AM, natasha agarwal
>>>> Hello Everyone,
>>>>
>>>> I was estimating the following model:
>>>>
>>>> probit expdum fdi if e>=0.50
>>>> margins, dydx(*)
>>>> probit expdum fdi if e<0.50
>>>> margins, dydx(*)
>>>>
>>>> Now I wanted to know whether the estimated Average partial effects of
>>>> fdi are statistically different across the two groups.
>>>>
>>>>
>>>> Many thanks
>>>> Natasha
>>>> *
>>>> *   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/
>>>>
>>>
>>>
>>>
>>> --
>>> Tirthankar Chakravarty
>>> tchakravarty@ucsd.edu
>>> tirthankar.chakravarty@gmail.com
>>>
>>
>>
>>
>> --
>> Tirthankar Chakravarty
>> tchakravarty@ucsd.edu
>> tirthankar.chakravarty@gmail.com
>>
>> *
>> *   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/
>

--
Tirthankar Chakravarty
tchakravarty@ucsd.edu
tirthankar.chakravarty@gmail.com

*
*   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/
```