istic. For example, some legal scholars argue that the Delay provisions are the only defenses with deterrent value [Coates 2000; Daines and Klausner 2001]. If managers also believe this,then the Delay subindex should also be the most important driver of the results.
Unfortunately, large standard errors, due in part to the sub stantial multicollinearity between the regressors, makes it difficult to construct a powerful test. None of the subindex coefficients are statistically significant in either specification, but many of the point estimates are economically large. In the end, we cannot precisely measure the relative importance of Delay or any other subindex. This is similar to the problem that occurred in the Q regressions of Table VIII. For example, in both Tables VIII and
XIII the coefficients on Voting suggest potentially enormous economic significance, but large standard errors prevent any meaningful statistical inference.In untabulated ests, we also included all 28 provisions from Table I as separate regressors in (3). Regressing raw returns on these 28 provisions plus the same controls as in Table XIII, we find that 16 of the coefficients are negative, and only one (Unequal Voting) is significant. (With this many regressors, we would expect one to appear "significant" just by chance.) Results for industry-adjusted returns are similar. These results highlight and magnify the lack of power in the su bindex regressions. In deed, many of the point estimates imply return effects above 20 basis points per month (2.4 percent per year), but are still far from being statistically significant. This result also suggests that the Democracy-minus-Dictatorship return differences are not driven by the presence or absence of any one provision.
istic ตัวอย่างเช่นบางนักวิชาการนักกฎหมายยืนยันว่าบทบัญญัติที่ล่าช้าในการเป็นเพียงการป้องกันที่มีค่ายับยั้ง Daines หากผู้บริหารยังเชื่อว่านี่แล้ว istic. For example, some legal scholars argue that the Delay provisions are the only defenses with deterrent value [Coates 2000; Daines and Klausner 2001]. If managers also believe this,then the Delay subindex should also be the most important driver of the results.
แต่ข้อผิดพลาดมาตรฐานขนาดใหญ่เนื่องจากในส่วนที่ย่อย ไม่มีค่าสัมประสิทธิ์ ในท้ายที่สุดเราไม่สามารถแม่นยำในการวัดความสำคัญของความล่าช้าหรือ นี้จะคล้ายกับปัญหาที่เกิดขึ้นในการถดถอย Unfortunately, large standard errors, due in part to the sub stantial multicollinearity between the regressors, makes it difficult to construct a powerful test. None of the subindex coefficients are statistically significant in either specification, but many of the point estimates are economically large. In the end, we cannot precisely measure the relative importance of Delay or any other subindex. This is similar to the problem that occurred in the Q regressions of Table VIII. For example, in both Tables VIII and
XIII the coefficients on Voting suggest potentially enormous economic significance, but large standard errors prevent any meaningful statistical inference.In untabulated ests, we also included all 28 provisions from Table I as separate regressors in (3). Regressing raw returns on these 28 provisions plus the same controls as in Table XIII, we find that 16 of the coefficients are negative, and only one (Unequal Voting) is significant. (With this many regressors, we would expect one to appear "significant" just by chance.) Results for industry-adjusted returns are similar. These results highlight and magnify the lack of power in the su bindex regressions. In deed, many of the point estimates imply return effects above 20 basis points per month (2.4 percent per year), but are still far from being statistically significant. This result also suggests that the Democracy-minus-Dictatorship return differences are not driven by the presence or absence of any one provision.
การแปล กรุณารอสักครู่..
