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 on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: st: Meaning of "." in Heckman Selection Models


From   Yuval Arbel <yuval.arbel@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: st: Meaning of "." in Heckman Selection Models
Date   Sat, 9 Jun 2012 23:50:37 +0300

Clifton, I just noted in the command you have X1 (with capital X) in
the main equation and x1 (with small x) in the selection equation. I'm
not sure this is the problem, but you can try.

If it doesn't work, and along the line Nick proposed, you can try to
run a -probit- model separately and to see what you get or if possible
to omit X1 from the selection equation.

BTW: Note that your lambda is marginally significant, which might be
another hint that something is problematic in the data structure. It
is very unlikely not to get a highly significant lambda in this
classical example of the heckman model.

On Sat, Jun 9, 2012 at 9:02 PM, Nick Cox <njcoxstata@gmail.com> wrote:
> In Stata an isolated . means missing. A standard error can't be
> estimated for X1, and so nothing else can be said. There is a chain of
> consequences for t, P and confidence intervals.
>
> The usual implication is that X1 is a dubious predictor, at least in
> the context of other predictors. Perhaps X1 does not vary, or there is
> a problem of collinearity. You should have explored your data
> previously and be able to relate to this to what you already know
> about the data, or be able to follow this up. Others should be able to
> add more about -heckman-, which I've never used.
>
> Nick
>
> On Sat, Jun 9, 2012 at 6:51 PM, Clifton Chow
> <clifton_chow@post.harvard.edu> wrote:
>
>> I just ran a two-step Heckman selection model and a peculiar result came out on the explanatory variable of interest in the selection equation.  Does someone know what a series of dots "." means in this context?  How should it be interpreted and reported?
>>
>> Here's the brief output from the results with X1 (a binary variable coded as 0 and 1) being my explanatory variable of interest.
>>
>> Heckman logwage X1 X2 X3 if female==1, select (x1 X2 X3 X4 X5) twostep
>>
>> LogWage Equation:
>>  X1 |  -.1290917   .0582952    -2.21   0.027    -.2433481   -.0148353
>>  X2 |  -.1528716   .0649873    -2.35   0.019    -.2802444   -.0254987
>>  X3 |   .1183843   .0684327     1.73   0.084    -.0157413    .2525098
>>
>> Selection Equation:
>> select
>>  X1 |   7.762731          .        .       .            .           .
>>  X2 |    .004973   .2220277     0.02   0.982    -.4301933    .4401393
>>  X3 |   .0188622    .238641     0.08   0.937    -.4488655    .4865898
>>  X4 |   .0196796   .1978019     0.10   0.921     -.368005    .4073641
>>  X5 |  -.0548543   .0817834    -0.67   0.502    -.2151467    .1054382
>>
>> mills      |
>>  lambda |  -.1344637   .0808789    -1.66   0.096    -.2929834    .0240561
>> -------------+----------------------------------------------------------------
>>      rho |   -0.38535
>>   sigma |  .34893481
>>  lambda | -.13446366   .0808789
>>
>
> *
> *   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/



-- 
Dr. Yuval Arbel
School of Business
Carmel Academic Center
4 Shaar Palmer Street,
Haifa 33031, Israel
e-mail1: yuval.arbel@carmel.ac.il
e-mail2: yuval.arbel@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/


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