Hi Mark,
I mean by control function the following step instead of using ivreg2
use http://fmwww.bc.edu/ec-p/data/wooldridge/mroz.dta
ivreg2 lwage exper expersq (educ=age kidslt6 kidsge6)
I use:
reg educ age kidslt6 kidsge6
predict ghat
reg lwage exper expersq educ ghat
The reason is that instead of lwage, I have binary variable called y so I run
probit y exper expersq educ ghat
So are we on the same line?
Thanks
nirina
On Mon, Dec 15, 2008 at 5:35 PM, Schaffer, Mark E <[email protected]> wrote:
> Nirina,
>
> You might want to check whether the Newey minimum chi-squared estimator
> implemented in -ivprobit- is actually what you're calling the control
> function estimator. -overid- reports the overid statistic for this
> estimator.
>
> Cheers,
> Mark
>
>> -----Original Message-----
>> From: [email protected]
>> [mailto:[email protected]] On Behalf Of Nirina F
>> Sent: 15 December 2008 21:59
>> To: [email protected]
>> Subject: Re: st: RE: exogeneity test for biprobit and control
>> function?
>>
>> Thank you very much for your response.
>> I actually need to use the control function not ivprobit so I was
>> wondering what would be the exogeneity test after -reg-.
>>
>> Any idea for the -biprobit- case too?
>> Nirina
>>
>> On Mon, Dec 15, 2008 at 4:23 PM, Schaffer, Mark E
>> <[email protected]> wrote:
>> > Nirina,
>> >
>> >> -----Original Message-----
>> >> From: [email protected]
>> >> [mailto:[email protected]] On Behalf Of Nirina F
>> >> Sent: 15 December 2008 19:41
>> >> To: [email protected]
>> >> Subject: st: exogeneity test for biprobit and control function?
>> >>
>> >> Dear all,
>> >>
>> >> Do you know what in STATA is the exogeneity test similar
>> to sargan or
>> >> Hansen's j (if I used robust) with ivreg2 that can be used with
>> >> biprobit and control function approach?
>> >>
>> >> The control function is for when I have binary dependent and
>> >> continuous independent variable.
>> >> Thank you,
>> >
>> > If you use a single-equation (limited information) framework and
>> > estimation your equation using -ivprobit-, you can use -overid-
>> > (downloadable from ssc-ideas in the usual way) to do a
>> post-estimation
>> > overidentification test.
>> >
>> > Hope this helps.
>> >
>> > Cheers,
>> > Mark (-overid- coauthor)
>> >
>> >>
>> >>
>> >> Nirina
>> >> *
>> >> * 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/
>> >>
>> >
>> >
>> > --
>> > Heriot-Watt University is a Scottish charity
>> > registered under charity number SC000278.
>> >
>> >
>> > *
>> > * 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/
>>
>
>
> --
> Heriot-Watt University is a Scottish charity
> registered under charity number SC000278.
>
>
> *
> * 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/
