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]

st: Interpretation of the estimates obtained by the -xtprobit- command


From   Yuval Arbel <yuval.arbel@gmail.com>
To   statalist <statalist@hsphsun2.harvard.edu>
Subject   st: Interpretation of the estimates obtained by the -xtprobit- command
Date   Thu, 29 Mar 2012 09:56:52 +0200

Dear Statalisters,

I'm attaching below the outcomes obtained for a sample of 8,909
pregnant women who gave birth to 25,052 infants

I have several questions:

1. Is there something equivalent to -dprobit- which permits direct
interpretation of the coefficients?

2. What is the reason -xtprobit- does not permit the fixed-effect option?

3. How can I test statistically the validity of the random-effect
between the constant term of each pregnant women and the variables?

4. I also ran the command -xtlogit- with the -fe- option (attached
below). The algorithm omitted many panels and I got the following
message:

note: multiple positive outcomes within groups encountered.
note: 7307 groups (18043 obs) dropped because of all positive or
      all negative outcomes.

I wonder what is the problem caused the algorithm to omit so many panels

I appreciate your assistance

Yuval

. xtprobit private2 hospitalization_num birth_by_operation mother_usa
mother_canada mother_england mother_ethiopia
> mother_switzerland mother_france mother_russia1 former_children1 muslim christian tourist1 immigrant pupil_in_yes
> hiva teacher

Fitting comparison model:

Iteration 0:   log likelihood = -8506.5487
Iteration 1:   log likelihood = -7708.4231
Iteration 2:   log likelihood = -7697.5432
Iteration 3:   log likelihood = -7697.5295
Iteration 4:   log likelihood = -7697.5295

Fitting full model:

rho =  0.0     log likelihood = -7697.5295
rho =  0.1     log likelihood =  -7628.272
rho =  0.2     log likelihood = -7632.6183

Iteration 0:   log likelihood =  -7628.272
Iteration 1:   log likelihood = -7624.7393
Iteration 2:   log likelihood = -7558.7882
Iteration 3:   log likelihood = -7558.4689
Iteration 4:   log likelihood = -7558.4685

Random-effects probit regression                Number of obs      =     25052
Group variable: index                           Number of groups   =      8908

Random effects u_i ~ Gaussian                   Obs per group: min =         1
                                                               avg =       2.8
                                                               max =        17

                                                Wald chi2(16)      =   1315.66
Log likelihood  = -7558.4685                    Prob > chi2        =    0.0000

------------------------------------------------------------------------------
    private2 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
hospitaliz~m |   .0129728   .0075312     1.72   0.085    -.0017881    .0277336
birth_by_o~n |   1.077024   .0383647    28.07   0.000      1.00183    1.152217
  mother_usa |   .8338842   .0495205    16.84   0.000     .7368257    .9309426
mother_can~a |   .8395324   .1346013     6.24   0.000     .5757187    1.103346
mother_eng~d |   .5693104   .0833725     6.83   0.000     .4059033    .7327175
mother_eth~a |  -.6634257   .2664923    -2.49   0.013    -1.185741   -.1411104
mother_swi~d |   .9164751   .1478142     6.20   0.000     .6267646    1.206186
mother_fra~e |  -.2771848   .1316733    -2.11   0.035    -.5352597   -.0191098
mother_rus~1 |  -.2880794   .1155616    -2.49   0.013     -.514576   -.0615828
former_chi~1 |   .0101862   .0046692     2.18   0.029     .0010346    .0193377
      muslim |  -.1308288   .0715678    -1.83   0.068    -.2710991    .0094414
   christian |  -1.073336   .5447206    -1.97   0.049    -2.140969   -.0057035
    tourist1 |    .444191   .0793486     5.60   0.000     .2886705    .5997114
   immigrant |   .1317313   .0493153     2.67   0.008     .0350752    .2283874
pupil_in_y~a |  -.0354636   .0331773    -1.07   0.285    -.1004899    .0295627
     teacher |  -.0135476   .0809299    -0.17   0.867    -.1721673    .1450721
       _cons |  -1.801318   .0438827   -41.05   0.000    -1.887327   -1.715309
-------------+----------------------------------------------------------------
    /lnsig2u |  -.9331074   .0921739                     -1.113765   -.7524499
-------------+----------------------------------------------------------------
     sigma_u |   .6271599   .0289039                      .5729926    .6864479
         rho |   .2822947   .0186748                      .2471697    .3202877
------------------------------------------------------------------------------
Likelihood-ratio test of rho=0: chibar2(01) =   278.12 Prob >= chibar2 = 0.000

. xtlogit private2 hospitalization_num birth_by_operation mother_usa
mother_canada mother_england mother_ethiopia m
> other_switzerland mother_france mother_russia1 former_children1 muslim christian tourist1 immigrant pupil_in_yesh
> iva teacher,fe
note: multiple positive outcomes within groups encountered.
note: 7307 groups (18043 obs) dropped because of all positive or
      all negative outcomes.

Iteration 0:   log likelihood = -2159.1289
Iteration 1:   log likelihood = -2120.0164
Iteration 2:   log likelihood = -2119.4414
Iteration 3:   log likelihood = -2119.3205
Iteration 4:   log likelihood =  -2119.296
Iteration 5:   log likelihood = -2119.2909
Iteration 6:   log likelihood = -2119.2898
Iteration 7:   log likelihood = -2119.2895
Iteration 8:   log likelihood = -2119.2894
Iteration 9:   log likelihood = -2119.2894

Conditional fixed-effects logistic regression   Number of obs      =      7009
Group variable: index                           Number of groups   =      1601

                                                Obs per group: min =         2
                                                               avg =       4.4
                                                               max =        17

                                                LR chi2(16)        =    702.85
Log likelihood  = -2119.2894                    Prob > chi2        =    0.0000

------------------------------------------------------------------------------
    private2 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
hospitaliz~m |   .0390685   .0175739     2.22   0.026     .0046242    .0735128
birth_by_o~n |   1.686039   .0937217    17.99   0.000     1.502348     1.86973
  mother_usa |   1.468732   .1319606    11.13   0.000     1.210094     1.72737
mother_can~a |     1.5773   .3641062     4.33   0.000     .8636647    2.290935
mother_eng~d |   1.090552   .2185774     4.99   0.000     .6621478    1.518955
mother_eth~a |  -1.215727   .7218827    -1.68   0.092    -2.630591    .1991374
mother_swi~d |   1.432616   .3794459     3.78   0.000     .6889158    2.176317
mother_fra~e |   -.502122   .3312722    -1.52   0.130    -1.151404    .1471596
mother_rus~1 |  -.3725172   .2961963    -1.26   0.209    -.9530514    .2080169
former_chi~1 |   .0169162   .0124888     1.35   0.176    -.0075614    .0413938
      muslim |  -.4356984   .1897871    -2.30   0.022    -.8076743   -.0637226
   christian |  -13.76266   558.0707    -0.02   0.980    -1107.561    1080.036
    tourist1 |   .6817092   .2114854     3.22   0.001     .2672056    1.096213
   immigrant |   .2331337   .1294273     1.80   0.072    -.0205391    .4868065
pupil_in_y~a |  -.1166618    .083074    -1.40   0.160    -.2794838    .0461603
     teacher |   .0254976   .1960549     0.13   0.897    -.3587629    .4097582
------------------------------------------------------------------------------



-- 
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