Table 11 Endogeneity test.

From: How effective is digital transformation? Heterogeneous insights from listed companies’ ESG performance

Variables

(1)

(2)

(3)

(4)

(5)

(6)

(7)

PSM

 

IV-regression

 

Heckman

 

Lag

Nearest-neighbor procedure

Kernel procedure

First stage

Second stage

First stage

Second stage

 

ESG

ESG

DT

ESG

DT-m

ESG

ESG

IV

  

0.0067*** (8.0520)

 

0.0051*** (5.3610)

  

DT

0.0535*** (10.0145)

0.0539*** (10.3687)

 

0.7578*** (5.4449)

 

0.0671*** (7.6761)

 

L.DT

      

0.0545*** (9.5064)

IMR

     

−0.0253* (−1.8734)

 

Constant

−0.1992 (−1.2853)

−0.2622* (−1.7535)

−1.6581*** (−9.3579)

0.6205** (2.0314)

−3.3486*** (−16.1466)

−0.2618* (−1.7497)

−0.5577*** (−3.3673)

Control variables

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Industry FE

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Year FE

Yes

Yes

Yes

Yes

Yes

Yes

Yes

LM Test

  

64.766***

    

Cragg–Donald Wald F test

  

64.835***

    

N

26,487

28,150

28,152

28,152

28,147

28,147

23,821

R-squared

0.1518

0.1518

–

–

–

0.1520

0.1583

  1. Columns (1) and (2) present the PSM results. Columns (3) and (4) present the results of IV 2SLS regressions. The LM statistic is 64.766, rejecting the under-identified hypothesis. The Wald F statistic is 64.835, which is greater than the threshold value of 10, indicating no significant weak instrumental variable. Columns (5) and (6) display the results of Heckman 2SLS regression. Finally, column (7) results show the effect of lag-DT. All continuous variables are winsorized at 1 and 99%. Columns (1), (2), (5), (6), and (7) report the T statistics in parentheses, while columns (3) and (4) report the Z statistics.
  2. *p < 0.1, **p < 0.05, ***p < 0.01.
Back to article page