This preview shows page 1. Sign up to view the full content.
Unformatted text preview: Aghion et al. (2003) (1) (2) 93% 93% Panel B: Including Both Actual and Random Delicensing
Percentage of times the actual delicensing-labor regulation interaction is significant at the 5% level 99% 98% Percentage of times the random delicensing-labor regulation interaction is significant at the 5% level 1% 7% Distribution of random delicensing years
delicensing Panel A: Actual Versus Random Delicensing
Percentage of times the actual delicensing-labor regulation interaction has a higher t-statistic than the random
delicensing-labor regulation interaction Notes: The table reports the results of Monte Carlo simulations. In each Monte Carlo simulation, one hundred random delicensing measures are created. Industries are allocated a random year of
delicensing and a random delicensing measure is created that equals one from the random year of delicensing onwards and zero otherwise. Random years of delicensing are based on the empirical
distribution of delicensing years. For each Monte Carlo simulation, the table reports statistical significance across one hundred regressions using the random delicensing measures. Standard errors are
adjusted for clustering by state × year of delicensing. 169 Openness and Economic Performance Development Economics, LSE Summer School 2007 Table 7: Falsification Checks on the Timing of Delicensing ...
View Full Document
This note was uploaded on 12/29/2011 for the course ECO 307 taught by Professor Dublin during the Spring '10 term at SUNY Stony Brook.
- Spring '10