Table 6

The endogeneity test by LIML estimator

Estimator: LIML and LPMInverted U-shaped relationship
LIMLLPM
1st stage2nd stage(Baseline regression)
DVsOptionsOptions2SuccessSuccess
LogTextlen0.013 (0. 035)0.042 (0.468)0.049*** (0.020)0.052*** (0.007)
LogVideos0.238*** (0.055)2.923*** (0.737)0.085*** (0.019)0.092*** (0.015)
LogPricerange0.527*** (0.011)6.492*** (0.143)0.004 (0.016)0.015*** (0.003)
LogTarget−0.148*** (0.014)−1.599*** (0.184)−0.055*** (0.008)−0.068*** (0.004)
LogDuration0.271*** (0.032)2.976*** (0.430)−0.130*** (0.016)−0.106*** (0.009)
Experience0.001*** (0.000)0.006*** (0.001)0.000*** (0.000)0.000*** (0.000)
SocialNetwork−0.052 (0.053)−0.868 (0.708)0.073*** (0.017)0.086*** (0.014)
Locations
IDs
ProjectType
BeginMonth
EndMonth
Options0.067*** (0.010)0.709*** (0.138)0.658** (0.254)0.064*** (0.012)
Options2−0.001** (0.000)−0.009* (0.005)−0.050** (0.020)−0.003*** (0.001)
Constant1.497*** (0.347)−13.940 (4.665)−1.141+ (0.673)0.670*** (0.087)
Observations8,9308,9308,9309,179
Summary of endogeneity test statistics
Under identification test
Kleibergen-Paap rk LM statistic15.72
p-Value0.000
Weak identification test
Cragg-Donald Wald F statistic11.38
Kleibergen-Paap rk Wald F statistic8.02
Stock-Yogo weak ID test critical values: 10% maximal IV size7.03

Note(s): (1) + p < 0.1, *p < 0.05, **p < 0.01, ***p < 0.001; (2) Observations Difference is caused by trimming the outliers of NoOfPicutres

or Create an Account

Close Modal
Close Modal