Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

Re: st: intreg cluster vs tobit cluster (one reports some missing std errors,


From   [email protected] (Jeff Pitblado, StataCorp LP)
To   [email protected]
Subject   Re: st: intreg cluster vs tobit cluster (one reports some missing std errors,
Date   Fri, 05 Nov 2010 10:52:23 -0500

Leandro Brufman <[email protected]> is getting a missing standard error
from -intreg- with -cluster()- using a dataset that is roughly 98% censored:

> I was running a Tobit model with clustered errors using -intreg
> varlist, clustered(clustervar)- as detailed in here
> (http://www.stata.com/support/faqs/stat/tobit.html). Sometimes the
> standard errors of a couple of coefficients appeared as missing. After
> reading a lot of useful things here in Statalist I couldn't find the
> problem behind those results (for example, I never had 1 observation
> per cluster, which was pointed out as a possible problem by Mark
> Schaefer, I fixed a scale problem in one variable that was driving a
> couple of missing std errors, but not all of them, etc.).
> 
> Just by chance I found one article of Woolridge (2006)
> (https://www.msu.edu/~ec/faculty/wooldridge/current%20research/clus1aea.pdf)
> At the end it says that you can run tobit with clustered errors by
> typing -tobit varlist, ll(0) cluster(clustervar)-
> I was just curious about that (I thought that tobit didn't allow
> cluster option). I run it and guess what.... the missing std errors
> dissapeared!
> 
> Below you'll see the results. Any ideas of why is this happening????

Leandro's output follows my signature.

We are having trouble reproducing this result.  If Leandro will contact me
privately with a copy of the dataset we can look into this more closely.

--Jeff
[email protected]

> *********** BEGIN EXAMPLE **************************
> 
> . intreg amt_mt2 amt_mt3 ip_vsam_ipolate_w cridum ccc*, cluster(cricode)
> 
> Fitting constant-only model:
> 
> Iteration 0:   log pseudolikelihood = -8848.0525
> Iteration 1:   log pseudolikelihood =  -1547.301
> Iteration 2:   log pseudolikelihood = -1425.1908
> Iteration 3:   log pseudolikelihood = -1407.5083
> Iteration 4:   log pseudolikelihood = -1407.1085
> Iteration 5:   log pseudolikelihood = -1407.1083
> Iteration 6:   log pseudolikelihood = -1407.1083
> 
> Fitting full model:
> 
> Iteration 0:   log pseudolikelihood = -8834.4116
> Iteration 1:   log pseudolikelihood = -1526.3341
> Iteration 2:   log pseudolikelihood = -1366.9358
> Iteration 3:   log pseudolikelihood = -1332.2446
> Iteration 4:   log pseudolikelihood = -1331.6778
> Iteration 5:   log pseudolikelihood = -1331.6745
> Iteration 6:   log pseudolikelihood = -1331.6745
> 
> Interval regression                               Number of obs   =       6757
>                                                   Wald chi2(11)   =     996.00
> Log pseudolikelihood = -1331.6745                 Prob > chi2     =     0.0000
> 
>                                (Std. Err. adjusted for 21 clusters in cricode)
> ------------------------------------------------------------------------------
>              |               Robust
>              |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
> -------------+----------------------------------------------------------------
> ip_vsam_ip~w |  -.0263045          .        .       .            .           .
>       cridum |  -156.2131    50.9088    -3.07   0.002    -255.9925   -56.43371
>         ccc1 |   219.9064   70.44264     3.12   0.002     81.84139    357.9715
>         ccc2 |   442.6939   147.2167     3.01   0.003     154.1545    731.2333
>         ccc3 |   97.37137   32.76476     2.97   0.003     33.15362    161.5891
>         ccc4 |   555.3243    185.262     3.00   0.003     192.2175    918.4311
>         ccc5 |   421.7124   142.3647     2.96   0.003     142.6827    700.7422
>         ccc6 |    269.146   87.67189     3.07   0.002     97.31224    440.9797
>         ccc8 |    96.5494   37.97573     2.54   0.011     22.11834    170.9804
>         ccc9 |   36.67169   13.64486     2.69   0.007     9.928247    63.41513
>        ccc10 |   179.9544   60.99526     2.95   0.003     60.40586    299.5029
>        ccc11 |   58.70021   28.58481     2.05   0.040      2.67502    114.7254
>        _cons |  -300.8123   94.71781    -3.18   0.001    -486.4558   -115.1688
> -------------+----------------------------------------------------------------
>     /lnsigma |   4.989588   .3018269    16.53   0.000     4.398018    5.581157
> -------------+----------------------------------------------------------------
>        sigma |   146.8758   44.33108                      81.28957    265.3786
> ------------------------------------------------------------------------------
> 
>   Observation summary:      6613  left-censored observations
>                              144     uncensored observations
>                                0 right-censored observations
>                                0       interval observations
> 
> . tobit amt_mt3 ip_vsam_ipolate_w cridum ccc*, ll(0) cluster(cricode)
> note: ccc7 dropped because of collinearity
> 
> Tobit regression                                  Number of obs   =       6757
>                                                   F(  12,   6745) =     287.30
>                                                   Prob > F        =     0.0000
> Log pseudolikelihood = -1331.6745                 Pseudo R2       =     0.0536
> 
>                                (Std. Err. adjusted for 21 clusters in cricode)
> ------------------------------------------------------------------------------
>              |               Robust
>      amt_mt3 |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
> -------------+----------------------------------------------------------------
> ip_vsam_ip~w |  -.0263045   .0088692    -2.97   0.003     -.043691    -.008918
>       cridum |  -156.2131    50.9064    -3.07   0.002    -256.0057   -56.42051
>         ccc1 |   219.9064   70.44073     3.12   0.002     81.82035    357.9925
>         ccc2 |   442.6939   147.2129     3.01   0.003     154.1102    731.2777
>         ccc3 |   97.37137   32.76389     2.97   0.003      33.1438    161.5989
>         ccc4 |   555.3243   185.2573     3.00   0.003     192.1616     918.487
>         ccc5 |   421.7124   142.3617     2.96   0.003     142.6385    700.7864
>         ccc6 |    269.146   87.66958     3.07   0.002     97.28593     441.006
>         ccc8 |    96.5494   37.97516     2.54   0.011     22.10609    170.9927
>         ccc9 |   36.67169   13.64448     2.69   0.007     9.924197    63.41918
>        ccc10 |   179.9544   60.99367     2.95   0.003     60.38752    299.5212
>        ccc11 |   58.70021   28.58445     2.05   0.040     2.665667    114.7348
>        _cons |  -300.8123   94.71551    -3.18   0.002    -486.4846     -115.14
> -------------+----------------------------------------------------------------
>       /sigma |   146.8758   44.33002                      59.97501    233.7767
> ------------------------------------------------------------------------------
>   Obs. summary:       6613  left-censored observations at amt_mt3<=0
>                        144     uncensored observations
>                          0 right-censored observations
> 
> 
> . compare amt_mt2 amt_mt3 // just to show you that the vars are ok,
> amt_mt2 is missing whenever amt_mt3==0, and both are equal otherwise.
> 
>                                         ---------- difference ----------
>                             count       minimum      average     maximum
> ------------------------------------------------------------------------
> amt_mt2=amt_mt3               152
>                        ----------
> jointly defined               152             0            0           0
> amt_mt2 missing only         6639
>                        ----------
> total                        6791
*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index