Dear All,
I will use xtpoisson command for the first time. I have read the Stata document of the command but I have still some questions. I would appreciate if you could clarify the points below:
1) What is the function of exposure variable? If I use "year" variable as exposure variable, does it mean I control the year effect in the regression model? I put my commands and results with and without YEAR dummies below . Could you please advise me which of the model is more reliable?
2) According to the stata document for xtpoisson command, xtpoission command can be used with re, fe and pa options. How can I select between the models? I know Hausman test is used to select between random and fixed effects, what about the comparison between pa and re&fe?
I will use xtpoisson command for the first time. I have read the Stata document of the command but I have still some questions. I would appreciate if you could clarify the points below:
1) What is the function of exposure variable? If I use "year" variable as exposure variable, does it mean I control the year effect in the regression model? I put my commands and results with and without YEAR dummies below . Could you please advise me which of the model is more reliable?
2) According to the stata document for xtpoisson command, xtpoission command can be used with re, fe and pa options. How can I select between the models? I know Hausman test is used to select between random and fixed effects, what about the comparison between pa and re&fe?
Code:
. xtpoisson GRI HDI EF FII FMI WGI TRADE, re exposure(YEAR)
Fitting Poisson model:
Iteration 0: log likelihood = -1575.8525
Iteration 1: log likelihood = -1572.6364
Iteration 2: log likelihood = -1572.6314
Iteration 3: log likelihood = -1572.6314
Fitting full model:
Iteration 0: log likelihood = -891.78467
Iteration 1: log likelihood = -813.44925
Iteration 2: log likelihood = -784.53822
Iteration 3: log likelihood = -780.32359
Iteration 4: log likelihood = -780.17044
Iteration 5: log likelihood = -780.17
Iteration 6: log likelihood = -780.17
Random-effects Poisson regression Number of obs = 168
Group variable: ID Number of groups = 24
Random effects u_i ~ Gamma Obs per group:
min = 7
avg = 7.0
max = 7
Wald chi2(6) = 298.45
Log likelihood = -780.17 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
GRI | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
HDI | 10.67069 1.254408 8.51 0.000 8.212095 13.12928
EF | .110718 .0529098 2.09 0.036 .0070167 .2144192
FII | 3.321821 .4692672 7.08 0.000 2.402074 4.241568
FMI | -1.182541 .3798706 -3.11 0.002 -1.927073 -.4380081
WGI | -.0070223 .0054438 -1.29 0.197 -.0176919 .0036474
TRADE | .0046995 .0023287 2.02 0.044 .0001353 .0092636
_cons | -12.85906 .975703 -13.18 0.000 -14.77141 -10.94672
ln(YEAR) | 1 (exposure)
-------------+----------------------------------------------------------------
/lnalpha | .5768572 .2882408 .0119156 1.141799
-------------+----------------------------------------------------------------
alpha | 1.780434 .5131938 1.011987 3.132398
------------------------------------------------------------------------------
LR test of alpha=0: chibar2(01) = 1584.92 Prob >= chibar2 = 0.000
. xtpoisson GRI HDI EF FII FMI WGI TRADE i.YEAR, re exposure(YEAR)
Fitting Poisson model:
Iteration 0: log likelihood = -1451.9488
Iteration 1: log likelihood = -1449.335
Iteration 2: log likelihood = -1449.3319
Iteration 3: log likelihood = -1449.3319
Fitting full model:
Iteration 0: log likelihood = -775.52875
Iteration 1: log likelihood = -735.45721
Iteration 2: log likelihood = -727.31961
Iteration 3: log likelihood = -726.81391
Iteration 4: log likelihood = -726.80864
Iteration 5: log likelihood = -726.80864
Random-effects Poisson regression Number of obs = 168
Group variable: ID Number of groups = 24
Random effects u_i ~ Gamma Obs per group:
min = 7
avg = 7.0
max = 7
Wald chi2(12) = 383.64
Log likelihood = -726.80864 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
GRI | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
HDI | -2.129105 1.810007 -1.18 0.239 -5.676654 1.418444
EF | .0816759 .0522113 1.56 0.118 -.0206564 .1840083
FII | 2.705611 .4744586 5.70 0.000 1.775689 3.635533
FMI | -.4729769 .3773934 -1.25 0.210 -1.212654 .2667007
WGI | -.000713 .0054791 -0.13 0.896 -.0114519 .010026
TRADE | .0001113 .0025765 0.04 0.966 -.0049386 .0051612
|
YEAR |
2011 | .2857912 .0524807 5.45 0.000 .1829309 .3886516
2012 | .4338611 .0538862 8.05 0.000 .3282461 .5394761
2013 | .5952082 .0597691 9.96 0.000 .478063 .7123534
2014 | .5961781 .0653963 9.12 0.000 .4680036 .7243525
2015 | .6499102 .0737901 8.81 0.000 .5052842 .7945362
2016 | .6037155 .0798463 7.56 0.000 .4472196 .7602114
|
_cons | -4.045653 1.281759 -3.16 0.002 -6.557855 -1.533451
ln(YEAR) | 1 (exposure)
-------------+----------------------------------------------------------------
/lnalpha | -.3780745 .3327322 -1.030218 .2740687
-------------+----------------------------------------------------------------
alpha | .6851795 .2279813 .3569293 1.315305
------------------------------------------------------------------------------
LR test of alpha=0: chibar2(01) = 1445.05 Prob >= chibar2 = 0.000
Code:
. xtpoisson GRI HDI EF FII FMI WGI TRADE, fe exposure(YEAR)
Iteration 0: log likelihood = -775.59867
Iteration 1: log likelihood = -599.96341
Iteration 2: log likelihood = -599.43074
Iteration 3: log likelihood = -599.43073
Conditional fixed-effects Poisson regression Number of obs = 168
Group variable: ID Number of groups = 24
Obs per group:
min = 7
avg = 7.0
max = 7
Wald chi2(6) = 337.40
Log likelihood = -599.43073 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
GRI | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
HDI | 12.79057 1.217706 10.50 0.000 10.40391 15.17723
EF | .1912765 .0506003 3.78 0.000 .0921017 .2904514
FII | 3.06273 .4725617 6.48 0.000 2.136526 3.988933
FMI | -1.283185 .3853587 -3.33 0.001 -2.038474 -.5278962
WGI | -.0009267 .0053867 -0.17 0.863 -.0114845 .0096311
TRADE | .0078161 .0022502 3.47 0.001 .0034058 .0122265
ln(YEAR) | 1 (exposure)
------------------------------------------------------------------------------
. xtpoisson GRI HDI EF FII FMI WGI TRADE i.YEAR, fe exposure(YEAR)
Iteration 0: log likelihood = -775.59867
Iteration 1: log likelihood = -562.0786
Iteration 2: log likelihood = -559.96972
Iteration 3: log likelihood = -559.96948
Iteration 4: log likelihood = -559.96948
Conditional fixed-effects Poisson regression Number of obs = 168
Group variable: ID Number of groups = 24
Obs per group:
min = 7
avg = 7.0
max = 7
Wald chi2(12) = 391.25
Log likelihood = -559.96948 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
GRI | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
HDI | -.4062424 2.476247 -0.16 0.870 -5.259596 4.447112
EF | .1910001 .0522478 3.66 0.000 .0885964 .2934038
FII | 2.509861 .4907356 5.11 0.000 1.548037 3.471685
FMI | -.7382897 .3941709 -1.87 0.061 -1.51085 .0342711
WGI | .0055388 .0057093 0.97 0.332 -.0056512 .0167288
TRADE | .0053887 .0025473 2.12 0.034 .0003961 .0103812
|
YEAR |
2011 | .2477582 .0535355 4.63 0.000 .1428306 .3526857
2012 | .3977779 .0568557 7.00 0.000 .2863428 .509213
2013 | .5578187 .0678905 8.22 0.000 .4247558 .6908817
2014 | .5495625 .0770526 7.13 0.000 .3985421 .7005829
2015 | .6285136 .0872209 7.21 0.000 .4575638 .7994635
2016 | .5883933 .0954913 6.16 0.000 .4012337 .7755528
ln(YEAR) | 1 (exposure)
------------------------------------------------------------------------------
Code:
. xtpoisson GRI HDI EF FII FMI WGI TRADE, pa exposure(YEAR)
Iteration 1: tolerance = 2.3385434
Iteration 2: tolerance = .20865524
Iteration 3: tolerance = .0478495
Iteration 4: tolerance = .01383021
Iteration 5: tolerance = .0044911
Iteration 6: tolerance = .00150734
Iteration 7: tolerance = .00051301
Iteration 8: tolerance = .00017535
Iteration 9: tolerance = .00006004
Iteration 10: tolerance = .00002057
Iteration 11: tolerance = 7.049e-06
Iteration 12: tolerance = 2.415e-06
Iteration 13: tolerance = 8.278e-07
GEE population-averaged model Number of obs = 168
Group variable: ID Number of groups = 24
Link: log Obs per group:
Family: Poisson min = 7
Correlation: exchangeable avg = 7.0
max = 7
Wald chi2(6) = 1246.81
Scale parameter: 1 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
GRI | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
HDI | 3.342809 .3234682 10.33 0.000 2.708822 3.976795
EF | -.0975146 .0121892 -8.00 0.000 -.1214049 -.0736242
FII | 3.550038 .1624921 21.85 0.000 3.23156 3.868517
FMI | -.7803941 .1158117 -6.74 0.000 -1.007381 -.5534074
WGI | -.0188183 .0015491 -12.15 0.000 -.0218546 -.0157821
TRADE | -.0091936 .0006503 -14.14 0.000 -.0104681 -.0079192
_cons | -5.883861 .2141177 -27.48 0.000 -6.303523 -5.464198
ln(YEAR) | 1 (exposure)
------------------------------------------------------------------------------
. xtpoisson GRI HDI EF FII FMI WGI TRADE i.YEAR, pa exposure(YEAR)
Iteration 1: tolerance = .48663477
Iteration 2: tolerance = .16379105
Iteration 3: tolerance = .08467556
Iteration 4: tolerance = .03321535
Iteration 5: tolerance = .01353904
Iteration 6: tolerance = .00538494
Iteration 7: tolerance = .00211855
Iteration 8: tolerance = .00082768
Iteration 9: tolerance = .0003224
Iteration 10: tolerance = .00012541
Iteration 11: tolerance = .00004875
Iteration 12: tolerance = .00001895
Iteration 13: tolerance = 7.364e-06
Iteration 14: tolerance = 2.862e-06
Iteration 15: tolerance = 1.112e-06
Iteration 16: tolerance = 4.321e-07
GEE population-averaged model Number of obs = 168
Group variable: ID Number of groups = 24
Link: log Obs per group:
Family: Poisson min = 7
Correlation: exchangeable avg = 7.0
max = 7
Wald chi2(12) = 1804.22
Scale parameter: 1 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
GRI | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
HDI | -1.32668 .3468089 -3.83 0.000 -2.006413 -.6469471
EF | .0179691 .0100962 1.78 0.075 -.0018191 .0377573
FII | 2.249052 .1565114 14.37 0.000 1.942295 2.555809
FMI | -.3056151 .1072055 -2.85 0.004 -.515734 -.0954962
WGI | -.0083042 .0015664 -5.30 0.000 -.0113742 -.0052342
TRADE | -.0101737 .0006594 -15.43 0.000 -.011466 -.0088814
|
YEAR |
2011 | .3365393 .0224792 14.97 0.000 .2924808 .3805977
2012 | .4809707 .0230003 20.91 0.000 .4358909 .5260506
2013 | .6241108 .0238508 26.17 0.000 .5773642 .6708574
2014 | .6194839 .0242414 25.55 0.000 .5719716 .6669962
2015 | .642498 .0253509 25.34 0.000 .5928112 .6921849
2016 | .5655204 .0259043 21.83 0.000 .5147488 .6162919
|
_cons | -3.225831 .2112304 -15.27 0.000 -3.639836 -2.811827
ln(YEAR) | 1 (exposure)
------------------------------------------------------------------------------
Comment