Bookmark and Share

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


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: RE: test for rho=0 in Heckman two-step procedure?


From   Tirthankar Chakravarty <tirthankar.chakravarty@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: RE: test for rho=0 in Heckman two-step procedure?
Date   Thu, 18 Feb 2010 05:48:54 +0530

See pg. 550 of Cameron & Trivedi's Microeconometrics (CUP, 2005) [CT]
for a discussion on how inference can be based on the inverse Mills
ratio (for which you do have a p-value reported in the output table).

Alternatively, you could try a bootstrap (recommended procedure in [CT]):
*********************************************
webuse womenwk, clear
bootstrap rho=e(rho), reps(1000): heckman wage education age,
select(married children education age) twostep
*********************************************

T

2010/2/18 Michael Boehm <michael.boehm1@gmail.com>:
> Probably I'm just blind, but I do really not find an option to do the
> "t-test" for rho=0. For example, if I do the command, there is no test
> statistic w.r.t. rho:
>
> . webuse womenwk
> . heckman wage educ age, select(married children educ age) twostep
>
> Heckman selection model -- two-step estimates   Number of obs      =      2000
> (regression model with sample selection)        Censored obs       =       657
>                                                Uncensored obs     =      1343
>
>                                                Wald chi2(2)       =    442.54
>                                                Prob > chi2        =    0.0000
>
> ------------------------------------------------------------------------------
>             |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
> -------------+----------------------------------------------------------------
> wage         |
>   education |   .9825259   .0538821    18.23   0.000     .8769189    1.088133
>
>         age |   .2118695   .0220511     9.61   0.000     .1686502    .2550888
>       _cons |   .7340391   1.248331     0.59   0.557    -1.712645    3.180723
> -------------+----------------------------------------------------------------
> select       |
>     married |   .4308575    .074208     5.81   0.000     .2854125    .5763025
>    children |   .4473249   .0287417    15.56   0.000     .3909922    .5036576
>   education |   .0583645   .0109742     5.32   0.000     .0368555    .0798735
>
>         age |   .0347211   .0042293     8.21   0.000     .0264318    .0430105
>       _cons |  -2.467365   .1925635   -12.81   0.000    -2.844782   -2.089948
> -------------+----------------------------------------------------------------
> mills        |
>      lambda |   4.001615   .6065388     6.60   0.000     2.812821     5.19041
> -------------+----------------------------------------------------------------
>         rho |    0.67284
>       sigma |  5.9473529
>      lambda |  4.0016155   .6065388
> ------------------------------------------------------------------------------
>
>
> On Wed, Feb 17, 2010 at 8:30 PM, Martin Weiss <martin.weiss1@gmx.de> wrote:
>>
>> <>
>>
>> All the options are arrayed in the help file, so what is the fuss about?
>>
>>
>> HTH
>> Martin
>>
>>
>> -----Original Message-----
>> From: owner-statalist@hsphsun2.harvard.edu
>> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Michael Boehm
>> Sent: Mittwoch, 17. Februar 2010 21:19
>> To: statalist@hsphsun2.harvard.edu
>> Subject: st: test for rho=0 in Heckman two-step procedure?
>>
>> Dear all,
>>
>> when I run the "heckman depvar [indepvars], select(varlist_s) twostep"
>> command I do not get a test for rho=0 (no correlation between errors
>> in the selection and the main equation) - contrary to the ML version
>> without the [twostep] option. I also do not find an option that
>> provides this test and couldn't find anything on a google search.
>>
>> Can someone help?
>>
>> Michael
>> *
>> *   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/
>



-- 
To every ω-consistent recursive class κ of formulae there correspond
recursive class signs r, such that neither v Gen r nor Neg(v Gen r)
belongs to Flg(κ) (where v is the free variable of r).

*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index