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: AW: Marginal-effect calculations after 'xtlogit, fe' ?


From   <severine_blaise@hotmail.com>
To   <statalist@hsphsun2.harvard.edu>
Subject   Re: st: AW: Marginal-effect calculations after 'xtlogit, fe' ?
Date   Wed, 30 Jun 2010 13:55:00 +1100

Dear Martin,

Thank you very much for your help.
I tried what is explained in the FAQ, but it yields only "zeros" :

do "C:\Users\toshiba\AppData\Local\Temp\STD00000000.tmp"
. xtlogit conflit txaide3 lpibt croiss service g txide lpop alimentpop eau, fe
note: multiple positive outcomes within groups encountered.
Iteration 0:   log likelihood = -165.51068
Iteration 1:   log likelihood = -127.98929
Iteration 2:   log likelihood = -126.88193
Iteration 3:   log likelihood = -126.87524
Iteration 4:   log likelihood = -126.87524
Conditional fixed-effects logistic regression Number of obs = 360 Group variable: pays Number of groups = 10 Obs per group: min = 36 avg = 36.0 max = 36 LR chi2(9) = 122.33 Log likelihood = -126.87524 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
conflit | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
txaide3 | -1.212498 .2253871 -5.38 0.000 -1.654249 -.7707475 lpibt | -1.229698 .5646258 -2.18 0.029 -2.336344 -.1230516 croiss | -2.28158 2.029903 -1.12 0.261 -6.260117 1.696956 service | .0166021 .0122666 1.35 0.176 -.0074399 .0406442 g | .0015161 .0426346 0.04 0.972 -.0820461 .0850784 txide | -2.193685 1.484768 -1.48 0.140 -5.103776 .7164069 lpop | 4.573573 .7922887 5.77 0.000 3.020716 6.126431
alimentpop |   .0025782   .0051149     0.50   0.614    -.0074468    .0126032
eau | -.0000184 .0000274 -0.67 0.503 -.0000722 .0000354
------------------------------------------------------------------------------

. mfx, predict(pu0)
Marginal effects after xtlogit
     y  = Pr(conflit|fixed effect is 0) (predict, pu0)
        =          1
------------------------------------------------------------------------------
variable |      dy/dx    Std. Err.     z    P>|z|  [    95% C.I.   ]      X
---------+--------------------------------------------------------------------
txaide3 | 0 0 . . 0 0 -2.52182 lpibt | 0 0 . . 0 0 5.8511 croiss | 0 0 . . 0 0 .025926 service | 0 0 . . 0 0 13.6778 g | 0 0 . . 0 0 14.6361 txide | 0 0 . . 0 0 .039854 lpop | 0 0 . . 0 0 16.2463 alimenp | 0 0 . . 0 0 3.78527 eau | 0 0 . . 0 0 21741.9
------------------------------------------------------------------------------
.
end of do-file


Do you have any idea of what is wrong ?

Thank you again.
Best regards,
Severine


--------------------------------------------------
From: "Martin Weiss" <martin.weiss1@gmx.de>
Sent: Wednesday, June 30, 2010 1:18 AM
To: <statalist@hsphsun2.harvard.edu>
Subject: st: AW: Marginal-effect calculations after  'xtlogit, fe' ?


<>


" xtlogit, conflit txaide lpibt croiss service g txide lpop alimentpop
eau,
re"


That first comma after -xtlogit- would make Stata think that "conflit" et
al. are options, and it would finally choke on "re"...

-h xtlogit postestimation- shows you what is available. Stata`s refusals
are
always grounded in proper theory in my experience, so sneaking your way
around them might be hazardous...


HTH
Martin

-----Ursprüngliche Nachricht-----
Von: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von
severine_blaise@hotmail.com
Gesendet: Dienstag, 29. Juni 2010 14:10
An: statalist@hsphsun2.harvard.edu
Betreff: st: Marginal-effect calculations after 'xtlogit, fe' ?

Hello all,

I understand that marginal effect calculations are only possible with the
default random effect of xtlogit, as follows :

xtlogit, conflit txaide lpibt croiss service g txide lpop alimentpop eau,
re
mfx compute, predict (pu0)

Does anyone know how to calculte such effects after a 'xtlgit, fe' ?
Thank you very much for your help.

Séverine


--------------------------------------------------
From: <severine_blaise@hotmail.com>
Sent: Friday, June 25, 2010 5:53 PM
To: <statalist@hsphsun2.harvard.edu>
Subject: Marginal-effect calculations and prediction tests with xtlogit
command

Hello,

I'm trying to calculate marginal effects after the estimation of a FE
logit model, and to obtain predictions... without success.
I'm working on the economic determinants of conflict onset and duration
in

10 African countries, therefore estimating with panel data, and
performing

tests on Stata 10.1. Below is my do file :

**Panel logit test on conflict onset/duration in African countries
log using Conflit, replace
set memory 18m
use ConflitAfrique
tsset pays temps, yearly
gen conflit1 = L.conflit
gen txaide = aide/pib
gen txaide3 = log(0.2*(L.txaide+L2.txaide+L3.txaide))
gen lpibt = log(pib/pop)
gen txpib = (pib-L.pib)/L.pib
gen croiss = L.txpib
gen txdette = dette/pib
gen txide = ide/pib
gen lpop = log(L.pop)
gen txaliment = (aliment-L.aliment)/L.aliment
gen txpop = (pop-L.pop)/L.pop
gen alimentpop = L.txaliment/L.txpop
gen leau = log(eau)
gen pibus3 = log(0.2*(L.pibus+L2.pibus+L3.pibus))
gen pibfr3 = log(0.2*(L.pibfr+L2.pibfr+L3.pibfr))

***Model of conflict onset
*Calculation of marginal effects
xtlogit conflit txaide lpibt croiss service g txide lpop alimentpop eau
if

conflit1==0, fe
mfx compute

/// RESULTS : default predict () is unsuitable for marginal-effect
calculation r(119);

*Prediction Test
xtlogit conflit conflit1 txaide lpibt croiss service txide lpop
alimentpop

ethnie eau, fe
predict confestime

/// RESULTS : (option pc1 assumed; conditional probability for single
outcome within group)
(20 missing values generated)
It seems to be working but I don't know how to display the results, as it
is the case with the command 'lstat' after a logit estimation.



Thank you very much in advance for your help.

Best Regards
Severine

****

Dr Severine BLAISE
Senior Lecturer in economics
University of New-Caledonia
Department of Law-Economics-Business
B.P. R4
98 851 Noumea CEDEX
New Caledonia

Web : http://www.severine-blaise.com/
Téléphone : (687) 26 69 31    Fax : (687) 26 69 30
 « Think green : print this message only if necessary »

*
*   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/


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