23. All of these additional variables are correlated with G (see Table III) and,
in prior studies, with either firm value or abnormal returns. See Lakonishok,
We estimate (3) separately for each month and then calculate the mean and time-series standard deviation of the 112 monthly estimates of the coefficients. Table XIII summarizes the results. The first two columns give the results, raw and industry-ad
justed, for the full sample of firms in each month with G as the key independent variable. In both regressions the average coefficient on G is negative but not significant. The point estimates are not small. For example, the point estimate for the coefficient on G in column (3) implies a lower return of approximately four bp per
month (=48 bp per year) for each additional point of G, but it would require estimates nearly twice as large before statistical significance would be reached.