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

# st: different p-values in mfx after xtprobit

 From ramesh To statalist@hsphsun2.harvard.edu Subject st: different p-values in mfx after xtprobit Date Thu, 12 May 2011 09:51:38 -0700 (PDT)

```Hi, I am using xtprobit model for my research. I want to find the marginal
effect after the xtprobit model. I used the "mfx compute, predict(pu0)"
command and got the results but my p-value is totally different and it turns
out 0.99 for almost all variable. I have attached the xt probit result and
its corresponding mfx herewith. Is there any suggest to get consistent
p-value in calculating the marginal effect?

Regards,
Ramesh Ghimire,
The University of Georgia
Athens, GA

------------------------------------------------------------------------------
flood |       Coef.         Std. Err.      z    P>|z|     [95% Conf.
Interval]
-------------+----------------------------------------------------------------
lnforest |  -.2013822   .0991885    -2.03    0.042     -.395788
-.0069763
lnpopden |   .8405052   .4208142     2.00   0.046     .0157245
1.665286
lnarea |   1.242506   .4060901     3.06     0.002     .4465838
2.038428
lnrain1 |   1.166008   .3712284     3.14     0.002     .4384142
1.893603
lnelevation |   .1471247    .553111     0.27    0.790    -.9369529
1.231202
lnlatitude |   .3573529   .4758066     0.75    0.453     -.575211
1.289917
lnrugged |  -.0959711   .4163809    -0.23   0.818    -.9120627
.7201205
dlandp |  -.0030963   .0064364    -0.48    0.630    -.0157114
.0095187
lndist_coast |   .0012429   .2458026     0.01   0.996    -.4805214
.4830072
subhumid |  -6.134987   17518.11    -0.00   1.000       -34341
34328.73
arid |     .3217398   .4151698     0.77    0.438     -.491978
1.135458
corruptionl |  -.2832393   .1430083    -1.98   0.048    -.5635303
-.0029482
lngdp2000l |   -.352413   .3048828    -1.16   0.248    -.9499723
.2451463
year |   .0412328   .0348107     1.18     0.236    -.0269949
.1094604
_cons |  -97.51084   69.47674    -1.40   0.160    -233.6828
38.66108
-------------+----------------------------------------------------------------
/lnsig2u |  -2.869749   1.663904                      -6.13094
.3914423
-------------+----------------------------------------------------------------
sigma_u |   .2381453   .1981254                      .0466319
1.216188
rho |   .0536694    .084508                      .0021698
.5966298
------------------------------------------------------------------------------
Likelihood-ratio test of rho=0: chibar2(01) =     0.51 Prob >= chibar2 =
0.238

. mfx compute, predict(pu0)

Marginal effects after xtprobit
y  = Pr(flood=1 assuming u_i=0) (predict, pu0)
=  .17457315
------------------------------------------------------------------------------
variable |      dy/dx         Std. Err.     z    P>|z|  [    95% C.I.   ]
X
---------+--------------------------------------------------------------------
lnforest |  -.0518309      34.252   -0.00     0.999  -67.1854  67.0817
9.86395
lnpopden |   .2163258      142.96    0.00    0.999  -279.978   280.41
4.36014
lnarea |   .3197912      211.33    0.00     0.999  -413.887  414.526
11.8466
lnrain1 |   .3001026      198.32    0.00     0.999  -388.405  389.005
5.51255
lnelev~n |   .0378664      25.024    0.00    0.999   -49.009  49.0847
5.77592
lnlati~e |   .0919741      60.781    0.00     0.999  -119.037  119.221
3.5787
lnrugged |  -.0247007      16.324   -0.00   0.999  -32.0187  31.9693
.02399
dlandp |  -.0007969      .52665   -0.00    0.999  -1.03301  1.03142
41.3952
l~_coast |   .0003199      .22067    0.00    0.999  -.432186  .432826
-2.36574
subhumid*|  -.2453962     .03779   -6.49   0.000  -.319465 -.171328
.040293
arid*|   .0874895      53.534    0.00      0.999  -104.836  105.011
.307692
corrup~l |   -.072899      48.175   -0.00     0.999  -94.4946  94.3488
4.48657
ln~2000l |  -.0907026      59.941   -0.00    0.999  -117.573  117.391
25.7931
year |   .0106123     7.01315    0.00     0.999  -13.7349  13.7561
1995.08
------------------------------------------------------------------------------
(*) dy/dx is for discrete change of dummy variable from 0 to 1

--
View this message in context: http://statalist.1588530.n2.nabble.com/different-p-values-in-mfx-after-xtprobit-tp6356445p6356445.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/
```