I am running an xtivewg2 command using some instrumental variables. I would like to reconfirm my interpretation of the post estimationn results:
1. Underidentification test (Anderson canon. corr. LM statistic): 58.187 (not significant)
My interpretation - Not underidentified
2. Weak identification test (Cragg-Donald Wald F statistic): 31.074 (ie. >10 and greater than all critical values)
My interpretation - the IVs are not weak
3. Sargan statistic (overidentification test of all instruments): 0.168 (significant)
My interpretation - The IVs are not overidentified
4. Endogeneity test of endogenous regressors: 5.340 (not significant)
My interpretation - the original regressor was endogenous
Please could you confirm my understanding? Thank you. I am pasting my code and results below:
Code:
. xtivreg2 Ln_EBIT_ROA Ln_Revenue Ln_LTD_to_Sales Ln_Intangible_Assets CoAge wGDPpc wCPI wDCF wEx > pgr wGDPgr wCons Ln_PS_RD (l1.Ln_GSD= l1.Ln_Indgrp_GSD_by_Year Ln_Int_exp) if CoAge>=0 & NATION > =="UNITED STATES" & NATIONCODE==840 & FSTS>=10 & GENERALINDUSTRYCLASSIFICATION ==1 & Year_<2020 > & Year_<YearInactive & Discr_GS_Rev!=1, fe endog (l1.Ln_GSD) Warning - singleton groups detected. 38 observation(s) not used. FIXED EFFECTS ESTIMATION ------------------------ Number of groups = 148 Obs per group: min = 2 avg = 5.9 max = 17 IV (2SLS) estimation -------------------- Estimates efficient for homoskedasticity only Statistics consistent for homoskedasticity only Number of obs = 870 F( 12, 710) = 5.52 Prob > F = 0.0000 Total (centered) SS = 244.9738206 Centered R2 = 0.0214 Total (uncentered) SS = 244.9738206 Uncentered R2 = 0.0214 Residual SS = 239.7376195 Root MSE = .5762 -------------------------------------------------------------------------------------- Ln_EBIT_ROA | Coef. Std. Err. z P>|z| [95% Conf. Interval] ---------------------+---------------------------------------------------------------- Ln_GSD | L1. | -.9935445 .3096328 -3.21 0.001 -1.600414 -.3866754 | Ln_Revenue | .5378527 .1230928 4.37 0.000 .2965952 .7791102 Ln_LTD_to_Sales | -.150372 .0329924 -4.56 0.000 -.215036 -.085708 Ln_Intangible_Assets | -.0759861 .041742 -1.82 0.069 -.157799 .0058268 CoAge | -.0324018 .0131353 -2.47 0.014 -.0581464 -.0066571 wGDPpc | .0000709 .0000283 2.51 0.012 .0000155 .0001262 wCPI | -.0039334 .0267595 -0.15 0.883 -.056381 .0485143 wDCF | 5.37e-14 1.50e-13 0.36 0.720 -2.40e-13 3.48e-13 wExpgr | .0117982 .0109984 1.07 0.283 -.0097582 .0333547 wGDPgr | -.0239097 .0322334 -0.74 0.458 -.0870861 .0392666 wCons | 1.56e-14 5.61e-14 0.28 0.782 -9.45e-14 1.26e-13 Ln_PS_RD | -.0379416 .0657001 -0.58 0.564 -.1667114 .0908283 -------------------------------------------------------------------------------------- Underidentification test (Anderson canon. corr. LM statistic): 58.187 Chi-sq(2) P-val = 0.0000 ------------------------------------------------------------------------------ Weak identification test (Cragg-Donald Wald F statistic): 31.074 Stock-Yogo weak ID test critical values: 10% maximal IV size 19.93 15% maximal IV size 11.59 20% maximal IV size 8.75 25% maximal IV size 7.25 Source: Stock-Yogo (2005). Reproduced by permission. ------------------------------------------------------------------------------ Sargan statistic (overidentification test of all instruments): 0.168 Chi-sq(1) P-val = 0.6822 -endog- option: Endogeneity test of endogenous regressors: 5.340 Chi-sq(1) P-val = 0.0208 Regressors tested: L.Ln_GSD ------------------------------------------------------------------------------ Instrumented: L.Ln_GSD Included instruments: Ln_Revenue Ln_LTD_to_Sales Ln_Intangible_Assets CoAge wGDPpc wCPI wDCF wExpgr wGDPgr wCons Ln_PS_RD Excluded instruments: L.Ln_Indgrp_GSD_by_Year Ln_Int_exp ------------------------------------------------------------------------------
0 Response to xtivreg2 and interpretation of post estimation results
Post a Comment