Quantcast

st: Heckman two-step estimator

classic Classic list List threaded Threaded
14 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

st: Heckman two-step estimator

Sridhar Telidevara
How would one go about estimating Heckman two-step estimator if both
the stages involve ordinal outcomes. Say the first stage has 5 levels
and the second stage has 3 levels. How would one calculate
inverse-mills ratio in the first step and incorporate them in the
second step. I would highly appreciate your help.

Thanks,

Sridhar
*
*   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/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

st: RE: Heckman two-step estimator

Martin Weiss-5

<>

Have you tried -ssc d cmp-?


HTH
Martin

-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of Sridhar
Telidevara
Sent: Sonntag, 14. Februar 2010 08:08
To: [hidden email]
Subject: st: Heckman two-step estimator

How would one go about estimating Heckman two-step estimator if both
the stages involve ordinal outcomes. Say the first stage has 5 levels
and the second stage has 3 levels. How would one calculate
inverse-mills ratio in the first step and incorporate them in the
second step. I would highly appreciate your help.

Thanks,

Sridhar
*
*   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/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: st: RE: Heckman two-step estimator

Sridhar Telidevara
In reply to this post by Sridhar Telidevara
Thank you Martin. I ran a bivariate seeming unrelated ordered probit
just to understand the command. It says ghk2setup() not found.
However, ghk2 is already installed and the software is up-to-date. I
am using stata11. I also ran an example given in the help. It says the
same. Searched web to learn more about ghk2setup() but no clue as to
what is happening.

Any thoughts?

Thanks,

Sridhar

On Sun, Feb 14, 2010 at 4:47 AM, Martin Weiss <[hidden email]> wrote:

>
> <>
>
> Have you tried -ssc d cmp-?
>
>
> HTH
> Martin
>
> -----Original Message-----
> From: [hidden email]
> [mailto:[hidden email]] On Behalf Of Sridhar
> Telidevara
> Sent: Sonntag, 14. Februar 2010 08:08
> To: [hidden email]
> Subject: st: Heckman two-step estimator
>
> How would one go about estimating Heckman two-step estimator if both
> the stages involve ordinal outcomes. Say the first stage has 5 levels
> and the second stage has 3 levels. How would one calculate
> inverse-mills ratio in the first step and incorporate them in the
> second step. I would highly appreciate your help.
>
> Thanks,
>
> Sridhar
> *
> *   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/
>

*
*   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/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

AW: st: RE: Heckman two-step estimator

Martin Weiss-5

<>

This example runs without a hitch for me:

*************
clear*
cmp setup
webuse laborsup
gen byte kids2 = kids + int(uniform()*3)
cmp (kids=fem_educ) (kids2=fem_educ), ind($cmp_oprobit $cmp_oprobit) nolr
tech(dfp) qui
*************


You may want to -ssc uninstall cmp- and then -ssc inst cmp-.

What does -which cmp- give you?


HTH
Martin

-----Ursprüngliche Nachricht-----
Von: [hidden email]
[mailto:[hidden email]] Im Auftrag von Sridhar
Telidevara
Gesendet: Sonntag, 14. Februar 2010 15:34
An: [hidden email]
Betreff: Re: st: RE: Heckman two-step estimator

Thank you Martin. I ran a bivariate seeming unrelated ordered probit
just to understand the command. It says ghk2setup() not found.
However, ghk2 is already installed and the software is up-to-date. I
am using stata11. I also ran an example given in the help. It says the
same. Searched web to learn more about ghk2setup() but no clue as to
what is happening.

Any thoughts?

Thanks,

Sridhar

On Sun, Feb 14, 2010 at 4:47 AM, Martin Weiss <[hidden email]> wrote:

>
> <>
>
> Have you tried -ssc d cmp-?
>
>
> HTH
> Martin
>
> -----Original Message-----
> From: [hidden email]
> [mailto:[hidden email]] On Behalf Of Sridhar
> Telidevara
> Sent: Sonntag, 14. Februar 2010 08:08
> To: [hidden email]
> Subject: st: Heckman two-step estimator
>
> How would one go about estimating Heckman two-step estimator if both
> the stages involve ordinal outcomes. Say the first stage has 5 levels
> and the second stage has 3 levels. How would one calculate
> inverse-mills ratio in the first step and incorporate them in the
> second step. I would highly appreciate your help.
>
> Thanks,
>
> Sridhar
> *
> *   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/
>

*
*   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/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: st: RE: Heckman two-step estimator

Sridhar Telidevara
In reply to this post by Sridhar Telidevara
I ran the same program that you sent me. It fits the individual models
but not the full model.

. clear*

.
. cmp setup

.
. webuse laborsup

.
. gen byte kids2 = kids + int(uniform()*3)

.
. cmp (kids=fem_educ) (kids2=fem_educ), ind($cmp_oprobit $cmp_oprobit) nolr

Fitting individual models as starting point for full model fit.

Iteration 0:   log likelihood = -800.76494
Iteration 1:   log likelihood = -800.48687
Iteration 2:   log likelihood = -800.48687

:::::results:::::::

Fitting full model.

               cmp_lnL():  3499  ghk2setup() not found
                 <istmt>:     -  function returned error
r(3499);

.
. tech(dfp) qui
unrecognized command:  tech
r(199);


Error 3499: The specified variable or function could not be found.
        In the case of a function, it was not already loaded, it is not
        in the libraries, and there is no .mo file with its name.


Sridhar


On Sun, Feb 14, 2010 at 9:41 AM, Martin Weiss <[hidden email]> wrote:

>
> <>
>
> This example runs without a hitch for me:
>
> *************
> clear*
> cmp setup
> webuse laborsup
> gen byte kids2 = kids + int(uniform()*3)
> cmp (kids=fem_educ) (kids2=fem_educ), ind($cmp_oprobit $cmp_oprobit) nolr
> tech(dfp) qui
> *************
>
>
> You may want to -ssc uninstall cmp- and then -ssc inst cmp-.
>
> What does -which cmp- give you?
>
>
> HTH
> Martin
>
> -----Ursprüngliche Nachricht-----
> Von: [hidden email]
> [mailto:[hidden email]] Im Auftrag von Sridhar
> Telidevara
> Gesendet: Sonntag, 14. Februar 2010 15:34
> An: [hidden email]
> Betreff: Re: st: RE: Heckman two-step estimator
>
> Thank you Martin. I ran a bivariate seeming unrelated ordered probit
> just to understand the command. It says ghk2setup() not found.
> However, ghk2 is already installed and the software is up-to-date. I
> am using stata11. I also ran an example given in the help. It says the
> same. Searched web to learn more about ghk2setup() but no clue as to
> what is happening.
>
> Any thoughts?
>
> Thanks,
>
> Sridhar
>
> On Sun, Feb 14, 2010 at 4:47 AM, Martin Weiss <[hidden email]> wrote:
>>
>> <>
>>
>> Have you tried -ssc d cmp-?
>>
>>
>> HTH
>> Martin
>>
>> -----Original Message-----
>> From: [hidden email]
>> [mailto:[hidden email]] On Behalf Of Sridhar
>> Telidevara
>> Sent: Sonntag, 14. Februar 2010 08:08
>> To: [hidden email]
>> Subject: st: Heckman two-step estimator
>>
>> How would one go about estimating Heckman two-step estimator if both
>> the stages involve ordinal outcomes. Say the first stage has 5 levels
>> and the second stage has 3 levels. How would one calculate
>> inverse-mills ratio in the first step and incorporate them in the
>> second step. I would highly appreciate your help.
>>
>> Thanks,
>>
>> Sridhar
>> *
>> *   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/
>>
>
> *
> *   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/
>

*
*   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/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

AW: st: RE: Heckman two-step estimator

Martin Weiss-5

<>


You still owe the list answers to the questions at the bottom of my previous
post, though...



HTH
Martin

-----Ursprüngliche Nachricht-----
Von: [hidden email]
[mailto:[hidden email]] Im Auftrag von Sridhar
Telidevara
Gesendet: Sonntag, 14. Februar 2010 16:07
An: [hidden email]
Betreff: Re: st: RE: Heckman two-step estimator

I ran the same program that you sent me. It fits the individual models
but not the full model.

. clear*

.
. cmp setup

.
. webuse laborsup

.
. gen byte kids2 = kids + int(uniform()*3)

.
. cmp (kids=fem_educ) (kids2=fem_educ), ind($cmp_oprobit $cmp_oprobit) nolr

Fitting individual models as starting point for full model fit.

Iteration 0:   log likelihood = -800.76494
Iteration 1:   log likelihood = -800.48687
Iteration 2:   log likelihood = -800.48687

:::::results:::::::

Fitting full model.

               cmp_lnL():  3499  ghk2setup() not found
                 <istmt>:     -  function returned error
r(3499);

.
. tech(dfp) qui
unrecognized command:  tech
r(199);


Error 3499: The specified variable or function could not be found.
        In the case of a function, it was not already loaded, it is not
        in the libraries, and there is no .mo file with its name.


Sridhar


On Sun, Feb 14, 2010 at 9:41 AM, Martin Weiss <[hidden email]> wrote:

>
> <>
>
> This example runs without a hitch for me:
>
> *************
> clear*
> cmp setup
> webuse laborsup
> gen byte kids2 = kids + int(uniform()*3)
> cmp (kids=fem_educ) (kids2=fem_educ), ind($cmp_oprobit $cmp_oprobit) nolr
> tech(dfp) qui
> *************
>
>
> You may want to -ssc uninstall cmp- and then -ssc inst cmp-.
>
> What does -which cmp- give you?
>
>
> HTH
> Martin
>
> -----Ursprüngliche Nachricht-----
> Von: [hidden email]
> [mailto:[hidden email]] Im Auftrag von Sridhar
> Telidevara
> Gesendet: Sonntag, 14. Februar 2010 15:34
> An: [hidden email]
> Betreff: Re: st: RE: Heckman two-step estimator
>
> Thank you Martin. I ran a bivariate seeming unrelated ordered probit
> just to understand the command. It says ghk2setup() not found.
> However, ghk2 is already installed and the software is up-to-date. I
> am using stata11. I also ran an example given in the help. It says the
> same. Searched web to learn more about ghk2setup() but no clue as to
> what is happening.
>
> Any thoughts?
>
> Thanks,
>
> Sridhar
>
> On Sun, Feb 14, 2010 at 4:47 AM, Martin Weiss <[hidden email]>
wrote:

>>
>> <>
>>
>> Have you tried -ssc d cmp-?
>>
>>
>> HTH
>> Martin
>>
>> -----Original Message-----
>> From: [hidden email]
>> [mailto:[hidden email]] On Behalf Of Sridhar
>> Telidevara
>> Sent: Sonntag, 14. Februar 2010 08:08
>> To: [hidden email]
>> Subject: st: Heckman two-step estimator
>>
>> How would one go about estimating Heckman two-step estimator if both
>> the stages involve ordinal outcomes. Say the first stage has 5 levels
>> and the second stage has 3 levels. How would one calculate
>> inverse-mills ratio in the first step and incorporate them in the
>> second step. I would highly appreciate your help.
>>
>> Thanks,
>>
>> Sridhar
>> *
>> *   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/
>>
>
> *
> *   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/
>

*
*   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/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: st: RE: Heckman two-step estimator

Sridhar Telidevara
In reply to this post by Sridhar Telidevara
I am sorry... I did reinstall cmp and

 -which cmp-

 which cmp
c:\ado\plus\c\cmp.ado
*! cmp 3.5.1 2 February 2010
*! David Roodman, Center for Global Development, Washington, DC, www.cgdev.org
*! Copyright David Roodman 2007-10. May be distributed free.

and the problem still remains even after reinstallation...

Fitting full model.

               cmp_lnL():  3499  ghk2setup() not found
                 <istmt>:     -  function returned error
r(3499);


Sridhar

On Sun, Feb 14, 2010 at 10:09 AM, Martin Weiss <[hidden email]> wrote:

>
> <>
>
>
> You still owe the list answers to the questions at the bottom of my previous
> post, though...
>
>
>
> HTH
> Martin
>
> -----Ursprüngliche Nachricht-----
> Von: [hidden email]
> [mailto:[hidden email]] Im Auftrag von Sridhar
> Telidevara
> Gesendet: Sonntag, 14. Februar 2010 16:07
> An: [hidden email]
> Betreff: Re: st: RE: Heckman two-step estimator
>
> I ran the same program that you sent me. It fits the individual models
> but not the full model.
>
> . clear*
>
> .
> . cmp setup
>
> .
> . webuse laborsup
>
> .
> . gen byte kids2 = kids + int(uniform()*3)
>
> .
> . cmp (kids=fem_educ) (kids2=fem_educ), ind($cmp_oprobit $cmp_oprobit) nolr
>
> Fitting individual models as starting point for full model fit.
>
> Iteration 0:   log likelihood = -800.76494
> Iteration 1:   log likelihood = -800.48687
> Iteration 2:   log likelihood = -800.48687
>
> :::::results:::::::
>
> Fitting full model.
>
>               cmp_lnL():  3499  ghk2setup() not found
>                 <istmt>:     -  function returned error
> r(3499);
>
> .
> . tech(dfp) qui
> unrecognized command:  tech
> r(199);
>
>
> Error 3499: The specified variable or function could not be found.
>        In the case of a function, it was not already loaded, it is not
>        in the libraries, and there is no .mo file with its name.
>
>
> Sridhar
>
>
> On Sun, Feb 14, 2010 at 9:41 AM, Martin Weiss <[hidden email]> wrote:
>>
>> <>
>>
>> This example runs without a hitch for me:
>>
>> *************
>> clear*
>> cmp setup
>> webuse laborsup
>> gen byte kids2 = kids + int(uniform()*3)
>> cmp (kids=fem_educ) (kids2=fem_educ), ind($cmp_oprobit $cmp_oprobit) nolr
>> tech(dfp) qui
>> *************
>>
>>
>> You may want to -ssc uninstall cmp- and then -ssc inst cmp-.
>>
>> What does -which cmp- give you?
>>
>>
>> HTH
>> Martin
>>
>> -----Ursprüngliche Nachricht-----
>> Von: [hidden email]
>> [mailto:[hidden email]] Im Auftrag von Sridhar
>> Telidevara
>> Gesendet: Sonntag, 14. Februar 2010 15:34
>> An: [hidden email]
>> Betreff: Re: st: RE: Heckman two-step estimator
>>
>> Thank you Martin. I ran a bivariate seeming unrelated ordered probit
>> just to understand the command. It says ghk2setup() not found.
>> However, ghk2 is already installed and the software is up-to-date. I
>> am using stata11. I also ran an example given in the help. It says the
>> same. Searched web to learn more about ghk2setup() but no clue as to
>> what is happening.
>>
>> Any thoughts?
>>
>> Thanks,
>>
>> Sridhar
>>
>> On Sun, Feb 14, 2010 at 4:47 AM, Martin Weiss <[hidden email]>
> wrote:
>>>
>>> <>
>>>
>>> Have you tried -ssc d cmp-?
>>>
>>>
>>> HTH
>>> Martin
>>>
>>> -----Original Message-----
>>> From: [hidden email]
>>> [mailto:[hidden email]] On Behalf Of Sridhar
>>> Telidevara
>>> Sent: Sonntag, 14. Februar 2010 08:08
>>> To: [hidden email]
>>> Subject: st: Heckman two-step estimator
>>>
>>> How would one go about estimating Heckman two-step estimator if both
>>> the stages involve ordinal outcomes. Say the first stage has 5 levels
>>> and the second stage has 3 levels. How would one calculate
>>> inverse-mills ratio in the first step and incorporate them in the
>>> second step. I would highly appreciate your help.
>>>
>>> Thanks,
>>>
>>> Sridhar
>>> *
>>> *   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/
>>>
>>
>> *
>> *   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/
>>
>
> *
> *   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/
>

*
*   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/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: st: RE: Heckman two-step estimator

Sridhar Telidevara
Thank you Martin.

Sridhar

On Sun, Feb 14, 2010 at 10:26 AM, Sridhar Telidevara
<[hidden email]> wrote:

> I am sorry... I did reinstall cmp and
>
>  -which cmp-
>
>  which cmp
> c:\ado\plus\c\cmp.ado
> *! cmp 3.5.1 2 February 2010
> *! David Roodman, Center for Global Development, Washington, DC, www.cgdev.org
> *! Copyright David Roodman 2007-10. May be distributed free.
>
> and the problem still remains even after reinstallation...
>
> Fitting full model.
>
>               cmp_lnL():  3499  ghk2setup() not found
>                 <istmt>:     -  function returned error
> r(3499);
>
>
> Sridhar
>
> On Sun, Feb 14, 2010 at 10:09 AM, Martin Weiss <[hidden email]> wrote:
>>
>> <>
>>
>>
>> You still owe the list answers to the questions at the bottom of my previous
>> post, though...
>>
>>
>>
>> HTH
>> Martin
>>
>> -----Ursprüngliche Nachricht-----
>> Von: [hidden email]
>> [mailto:[hidden email]] Im Auftrag von Sridhar
>> Telidevara
>> Gesendet: Sonntag, 14. Februar 2010 16:07
>> An: [hidden email]
>> Betreff: Re: st: RE: Heckman two-step estimator
>>
>> I ran the same program that you sent me. It fits the individual models
>> but not the full model.
>>
>> . clear*
>>
>> .
>> . cmp setup
>>
>> .
>> . webuse laborsup
>>
>> .
>> . gen byte kids2 = kids + int(uniform()*3)
>>
>> .
>> . cmp (kids=fem_educ) (kids2=fem_educ), ind($cmp_oprobit $cmp_oprobit) nolr
>>
>> Fitting individual models as starting point for full model fit.
>>
>> Iteration 0:   log likelihood = -800.76494
>> Iteration 1:   log likelihood = -800.48687
>> Iteration 2:   log likelihood = -800.48687
>>
>> :::::results:::::::
>>
>> Fitting full model.
>>
>>               cmp_lnL():  3499  ghk2setup() not found
>>                 <istmt>:     -  function returned error
>> r(3499);
>>
>> .
>> . tech(dfp) qui
>> unrecognized command:  tech
>> r(199);
>>
>>
>> Error 3499: The specified variable or function could not be found.
>>        In the case of a function, it was not already loaded, it is not
>>        in the libraries, and there is no .mo file with its name.
>>
>>
>> Sridhar
>>
>>
>> On Sun, Feb 14, 2010 at 9:41 AM, Martin Weiss <[hidden email]> wrote:
>>>
>>> <>
>>>
>>> This example runs without a hitch for me:
>>>
>>> *************
>>> clear*
>>> cmp setup
>>> webuse laborsup
>>> gen byte kids2 = kids + int(uniform()*3)
>>> cmp (kids=fem_educ) (kids2=fem_educ), ind($cmp_oprobit $cmp_oprobit) nolr
>>> tech(dfp) qui
>>> *************
>>>
>>>
>>> You may want to -ssc uninstall cmp- and then -ssc inst cmp-.
>>>
>>> What does -which cmp- give you?
>>>
>>>
>>> HTH
>>> Martin
>>>
>>> -----Ursprüngliche Nachricht-----
>>> Von: [hidden email]
>>> [mailto:[hidden email]] Im Auftrag von Sridhar
>>> Telidevara
>>> Gesendet: Sonntag, 14. Februar 2010 15:34
>>> An: [hidden email]
>>> Betreff: Re: st: RE: Heckman two-step estimator
>>>
>>> Thank you Martin. I ran a bivariate seeming unrelated ordered probit
>>> just to understand the command. It says ghk2setup() not found.
>>> However, ghk2 is already installed and the software is up-to-date. I
>>> am using stata11. I also ran an example given in the help. It says the
>>> same. Searched web to learn more about ghk2setup() but no clue as to
>>> what is happening.
>>>
>>> Any thoughts?
>>>
>>> Thanks,
>>>
>>> Sridhar
>>>
>>> On Sun, Feb 14, 2010 at 4:47 AM, Martin Weiss <[hidden email]>
>> wrote:
>>>>
>>>> <>
>>>>
>>>> Have you tried -ssc d cmp-?
>>>>
>>>>
>>>> HTH
>>>> Martin
>>>>
>>>> -----Original Message-----
>>>> From: [hidden email]
>>>> [mailto:[hidden email]] On Behalf Of Sridhar
>>>> Telidevara
>>>> Sent: Sonntag, 14. Februar 2010 08:08
>>>> To: [hidden email]
>>>> Subject: st: Heckman two-step estimator
>>>>
>>>> How would one go about estimating Heckman two-step estimator if both
>>>> the stages involve ordinal outcomes. Say the first stage has 5 levels
>>>> and the second stage has 3 levels. How would one calculate
>>>> inverse-mills ratio in the first step and incorporate them in the
>>>> second step. I would highly appreciate your help.
>>>>
>>>> Thanks,
>>>>
>>>> Sridhar
>>>> *
>>>> *   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/
>>>>
>>>
>>> *
>>> *   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/
>>>
>>
>> *
>> *   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/
>>
>

*
*   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/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: st: RE: Heckman two-step estimator

Rober
Dear Sridhar

I am experiencing the same problem you reported here.
Have you made any progress trying to fix this?

Any hints would be appreciated

Rober
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: st: RE: Heckman two-step estimator

Siri
Dear Rober, and other statalisters

I am also having problems with cmp. The message I receive when it is suppose to fit the full model is:

 <istmt>:  3499  cmp_lnL() not found

 So if anyone makes any progress on this problem, I would truly appreciate if it is shared on this thread.

Siri
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

RE: Heckman two-step estimator

Martin Weiss-5

<>

Just to be sure, Siri, are you picking up this thread dating back all the
way to Feb: http://www.stata.com/statalist/archive/2010-02/msg00629.html?
Maybe you want to consider a change in your policy to delete all the stuff
you are replying to. Makes it hard to view your post in context...


HTH
Martin

-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of Siri
Sent: Mittwoch, 1. September 2010 23:13
To: [hidden email]
Subject: Re: st: RE: Heckman two-step estimator

Dear Rober, and other statalisters

I am also having problems with cmp. The message I receive when it is suppose
to fit the full model is:

 <istmt>:  3499  cmp_lnL() not found

 So if anyone makes any progress on this problem, I would truly appreciate
if it is shared on this thread.

Siri
--
View this message in context:
http://statalist.1588530.n2.nabble.com/st-Heckman-two-step-estimator-tp45694
36p5488841.html
Sent from the Statalist mailing list archive at Nabble.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/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

RE: Heckman two-step estimator

Maarten buis
In reply to this post by Siri
--- On Wed, 1/9/10, Siri wrote:
> I am also having problems with cmp. The message I receive
> when it is suppose to fit the full model is:
>
>  <istmt>:  3499  cmp_lnL() not found

-cmp- uses Mata functions, and the error you got means that
Stata could not find that particular Mata function. Stata does
not automatically find new Mata functions when you just
installed an new package. The easiest solution is to close down
Stata, start it again, and try to use -cmp- again.

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/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

RE: Heckman two-step estimator

Siri
Thank you Maarten.
 
For some reason I actually had to reinstall Stata, but now it works perfectly
 
Siri
 

Date: Thu, 2 Sep 2010 03:05:45 -0700
From: [hidden email]
To: [hidden email]
Subject: RE: Heckman two-step estimator

--- On Wed, 1/9/10, Siri wrote:
> I am also having problems with cmp. The message I receive
> when it is suppose to fit the full model is:
>
>  <istmt>:  3499  cmp_lnL() not found

-cmp- uses Mata functions, and the error you got means that
Stata could not find that particular Mata function. Stata does
not automatically find new Mata functions when you just
installed an new package. The easiest solution is to close down
Stata, start it again, and try to use -cmp- again.

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/



View message @ http://statalist.1588530.n2.nabble.com/st-Heckman-two-step-estimator-tp4569436p5490523.html
To unsubscribe from st: Heckman two-step estimator, click here.

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Testing for overidentifying restrictions iv-oprobit

Siri
In reply to this post by Maarten buis
Dear statalisters

I am using cmp to estimate an ordered probit with an endogenous variable. cmp does this for me in a very simple manner, but I havent found a way to test for overidentifying restrictions. Do you know if there is any way to do this? My command is of the following type:

cmp (endogenous_var = incl_instruments excl_instruments) (dependent_ordered_var = endogenous_var incl_instruments), ind($cmp_cont $cmp_oprobit)

Any advices would be greatly appreciated

Siri    
*
*   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/
Loading...