Table 13

Panel A – Heckman 2SLS selection bias test

Variables(1)(2)(3)(4)(5)
GAPU_medianTOBINQMKTBKBMEETLNBMEET
1st stage2nd stage2nd stage2nd stage2nd stage
GAPU −0.005***−0.000***−0.001***−0.001**
 (−3.616)(−3.152)(−3.248)(−2.511)
CRG1.876***    
(57.725)    
IMR 0.0130.000**0.011***0.019***
 (0.769)(2.327)(2.770)(4.451)
ControlsYesYesYesYesYes
Constant−0.625**16.733***2.406***1.083***0.646***
(−2.238)(22.073)(308.711)(14.222)(8.468)
IndustryYesYesYesYesYes
YearYesYesYesYesYes
Observations37,20037,20037,20037,20037,200
Adjusted-R2/Pseudo R20.17330.41650.89040.06370.2105
Panel B – Heckman 2SLS selection bias test
Variables(1)(2)(3)(4)(5)
GAPS_medianTOBINQMKTBKBMEETLNBMEET
1st stage2nd stage2nd stage2nd stage2nd stage
GAPS −0.002***−0.000**−0.001***−0.001***
 (−2.798)(−2.213)(−3.617)(−4.142)
CRG1.239***    
(40.416)    
IMR 0.063***0.001***0.018***0.026***
 (2.679)(3.697)(3.205)(4.764)
ControlsYesYesYesYesYes
Constant−1.709***16.589***2.404***1.056***0.617***
(−7.203)(21.646)(305.349)(13.586)(7.974)
IndustryYesYesYesYesYes
YearYesYesYesYesYes
Observations37,20037,20037,20037,20037,125
Adjusted-R2/Pseudo R20.12920.41650.89040.06370.2106

Note(s): Table 13 Panel A (Dependent = GAPU) and Panel B (Dependent = GAPS) report the Heckman 2SLS Selection Bias Test. t-statistics in parentheses. Standard error clustered based on industry and year. *p < 0.10, **p < 0.05, ***p < 0.01. Refer to  Appendix for variable definitions. Column (1) is Logit regression, columns (2), (3), and (5) are OLS regressions, and column (4) is Poisson regression

Source(s): Authors’ own calculation

or Create an Account

Close Modal
Close Modal