For this control purpose, the BHAR approach with the control firm method advocated by Barber and Lyon (1997) is adopted in this study.
Average BHAR, the excess of BHR's of restating firms over those of the matching non-restating firms, are not significant in any single month and holding period following the announcement of earnings restatements.
Third, we show that a firm's three-year BHAR is positively related to the logarithm of the time between its IPO and first SEO.
The BHAR is over a three-year holding period following the SEO issue.
Barber and Lyon (1997) recommend the use of BHAR based on a size- and BM-matched firm approach as this eliminates the biases in test statistics designed to detect long-run abnormal returns.
We exclude the initial return and measure BHAR from the day after IPOs.
If we replace the Year -1 CAR with the Year -1 BHAR, the results remain qualitatively the same.
For BHARs, we subtract control-firm holding-period returns from sample-firm holding-period returns.
We define D_BHAR as the difference between the BHAR of the reverse split firm and the mean BHAR of the control group of otherwise similar firms that did not split their shares explained in Section I.
Therefore, D_BHAR measures the portion of BHAR that is attributable to the reverse stock split and is a more accurate measure of the effect of reverse splits on share prices.
More importantly, the differences between exercises that are dividend motivated and those that are not are significant, with Type I errors of less than 1% using BHARs
, CTARs, raw, market-adjusted returns, and BHAR
High-growth firms also have lower returns when using BHARs
as the metric instead of annualized returns as noted in Table IX.