Andrew Musau , Richard Williams
I run the same model with two approaches , it suppose to give me the same results but in my analysis there are slightly differences , so from your opinion which model is better and why??
Code:
Augmented Mean Group estimator (Bond & Eberhardt, 2009; Eberhardt & Teal, 2010) Common dynamic process included as additional regressor All coefficients represent averages across groups (group variable: CountryID) Coefficient averages computed as outlier-robust means (using rreg) Mean Group type estimation Number of obs = 618 AMG Wald chi2(8) = 27.90 Prob > chi2 = 0.0005 Y Coefficient Std. err. z P>z [95% conf. interval] Indep1_P -.0115727 .0230853 -0.50 0.616 -.0568191 .0336737 Indep1_N .022238 .0302148 0.74 0.462 -.0369818 .0814578 Indep2_P .0401797 .026852 1.50 0.135 -.0124494 .0928087 Indep2_N -.0263027 .0129571 -2.03 0.042 -.0516983 -.0009072 Indep3_P -.0145813 .0174016 -0.84 0.402 -.0486879 .0195253 indep3_N -.035766 .0368973 -0.97 0.332 -.1080835 .0365514 c_d_p .1165165 .09198 1.27 0.205 -.063761 .2967939 trend .0000876 .0002515 0.35 0.728 -.0004053 .0005805 _cons -.030048 .0173433 -1.73 0.083 -.0640403 .0039444 Root Mean Squared Error (sigma): 0.0442 (RMSE uses residuals from group-specific regressions: unaffected by 'robust'). Variable c_d_p refers to the common dynamic process. Variable trend refers to the group-specific linear trend terms. Share of group-specific trends significant at 5% level: 0.091 (= 2 trends)
Code:
ugmented Mean Group estimator (Bond & Eberhardt, 2009; Eberhardt & Teal, 2010) Common dynamic process included as additional regressor All coefficients represent averages across groups (group variable: CountryID) Coefficient averages computed as outlier-robust means (using rreg) Mean Group type estimation Number of obs = 618 AMG Wald chi2(8) = 28.00 Prob > chi2 = 0.0005 Y Coefficient Std. err. z P>z [95% conf. interval] Indep1 .022238 .0302148 0.74 0.462 -.0369818 .0814578 Indep2 -.0288838 .0126964 -2.27 0.023 -.0537682 -.0039993 Indep3 .0060371 .0138635 0.44 0.663 -.0211349 .0332091 inter_indep1P -.0223807 .0506669 -0.44 0.659 -.1216859 .0769246 inter_indep2P .0262421 .0178814 1.47 0.142 -.0088047 .0612889 inter_indep3P -.0230171 .0289045 -0.80 0.426 -.0796689 .0336347 c_d_p .1165165 .09198 1.27 0.205 -.0637609 .2967939 trend .0000876 .0002515 0.35 0.728 -.0004053 .0005805 _cons -.0048551 .009756 -0.50 0.619 -.0239765 .0142663 Root Mean Squared Error (sigma): 0.0442 (RMSE uses residuals from group-specific regressions: unaffected by 'robust'). Variable c_d_p refers to the common dynamic process. Variable trend refers to the group-specific linear trend terms. Share of group-specific trends significant at 5% level: 0.091 (= 2 trends)
0 Response to Asymmetric approaches and panel data ??
Post a Comment