I have a question concerning the a logistic regression with panel data. I have a panel dataset containing only 61 observations and 12 groups. My dependent variable is whether sexual violence has happened in a conflict or not. I ran a regression using xtlogit and everything worked perfectly fine, but I discovered that the probability of the LR test at the bottom is 1. Now, I do not know whether I have to change anything or what to do with it? I looked it up in the Stata Manual and it says, that this test compares the pooled estimator (logit) with the panel estimator, but what does that mean?
I am working with version 15 in Stata.
This is my code:
Code:
xtlogit sexual_violence_bin women_share GenderGap P3_Human_Rights polity loggdp Population i.conflict_activity year_count region, re Fitting comparison model: Iteration 0: log likelihood = -33.961598 Iteration 1: log likelihood = -17.415634 Iteration 2: log likelihood = -14.828786 Iteration 3: log likelihood = -12.420276 Iteration 4: log likelihood = -11.272802 Iteration 5: log likelihood = -11.182178 Iteration 6: log likelihood = -11.180721 Iteration 7: log likelihood = -11.18072 Fitting full model: tau = 0.0 log likelihood = -11.18072 tau = 0.1 log likelihood = -11.332501 Iteration 0: log likelihood = -11.332501 Iteration 1: log likelihood = -11.181432 Iteration 2: log likelihood = -11.180724 Iteration 3: log likelihood = -11.18072 Iteration 4: log likelihood = -11.18072 Random-effects logistic regression Number of obs = 53 Group variable: missionid Number of groups = 12 Random effects u_i ~ Gaussian Obs per group: min = 2 avg = 4.4 max = 6 Integration method: mvaghermite Integration pts. = 12 Wald chi2(10) = 5.58 Log likelihood = -11.18072 Prob > chi2 = 0.8496 ------------------------------------------------------------------------------------- sexual_violence_bin | Coef. Std. Err. z P>|z| [95% Conf. Interval] --------------------+---------------------------------------------------------------- women_share | -34.39814 47.97645 -0.72 0.473 -128.4303 59.63398 GenderGap | 40.76391 25.25293 1.61 0.106 -8.730913 90.25874 P3_Human_Rights | 4.845226 2.226682 2.18 0.030 .4810104 9.209442 polity | -6.039683 3.102269 -1.95 0.052 -12.12002 .0406531 loggdp | -1.842643 1.797294 -1.03 0.305 -5.365273 1.679988 Population | 1.49e-08 2.06e-08 0.72 0.471 -2.55e-08 5.52e-08 | conflict_activity | 2 | 2.31902 3.866796 0.60 0.549 -5.259761 9.897802 3 | 6.933563 4.103001 1.69 0.091 -1.108171 14.9753 | year_count | -.8181246 .4552427 -1.80 0.072 -1.710384 .0741347 region | -4.842108 2.928383 -1.65 0.098 -10.58163 .8974175 _cons | .0909244 26.32624 0.00 0.997 -51.50757 51.68941 --------------------+---------------------------------------------------------------- /lnsig2u | -17.75879 1554.457 -3064.439 3028.921 --------------------+---------------------------------------------------------------- sigma_u | .0001392 .1082125 0 . rho | 5.89e-09 9.16e-06 0 . ------------------------------------------------------------------------------------- LR test of rho=0: chibar2(01) = 0.00 Prob >= chibar2 = 1.000
0 Response to Panel-Data Problems with the xtlogit command
Post a Comment