0032 0075 b fe 0050 mod 0014 0013 n 1718 high 0030

This preview shows page 20 - 23 out of 31 pages.

- 0.032 - 0.075% B FE \ - 0.050% Mod - 0.014 - 0.013 N = 1,718 High - 0.030 0.027 Interval - 2 Low - 0.008 - 0.038 - 0.050% B FE \ - 0.025% Mod - 0.010 - 0.009 N = 2,388 High - 0.031 0.012 Interval - 1 Low - 0.014 - 0.038 - 0.025% B FE \ 0 Mod - 0.017 - 0.015 N = 936 High - 0.015 - 0.001 Interval 0 Low - 0.003 - 0.038 FE = 0 Mod - 0.006 - 0.003 N = 10,053 High - 0.027 0.022 Interval 1 Low 0.001 - 0.041 0 \ FE B 0.025% Mod 0.001 0.001 N = 2,664 High - 0.019 0.013 20 L. Rees, W. Thomas 123
Image of page 20

Subscribe to view the full document.

Table 4 continued Panel A: Forecast error intervals D Dispersion ST Revision Interval 2 Low 0.003 - 0.030 0.025% \ FE B 0.050% Mod 0.009 0.008 N = 6,512 High - 0.012 0.019 Interval 3 Low 0.011 - 0.019 0.050% \ FE B 0.075% Mod 0.015 0.015 N = 5,025 High 0.003 0.032 Interval 4 Low 0.015 - 0.014 0.075% \ FE B 0.100% Mod 0.018 0.018 N = 3,720 High 0.010 0.033 Interval 5 Low 0.020 - 0.020 0.100% \ FE B 0.125% Mod 0.023 0.024 N = 2,901 High 0.015 0.038 Interva 6 Low 0.023 - 0.010 0.125% \ FE B 0.150% Mod 0.022 0.024 N = 2,353 High 0.020 0.046 Interval 7 Low 0.026 - 0.005 0.150% \ FE B 0.175% Mod 0.026 0.026 N = 1,752 High 0.018 0.040 Interval 8 Low 0.027 0.000 0.175% \ FE B 0.200% Mod 0.026 0.027 N = 1,524 High 0.028 0.047 Interval 9 Low 0.029 - 0.007 0.200% \ FE B 0.225% Mod 0.031 0.027 N = 1,237 High 0.019 0.055 Interval 10 Low 0.033 0.001 0.225% \ FE Mod 0.033 0.034 N = 9,060 High 0.037 0.056 Panel B: Summary of the difference in mean announcement period returns of firms in the High and Low groups in each forecast error interval ( N = 21 FE intervals in Panel A) Mean ( t -stat.) Mean ( t -stat.) Difference (High–Low) - 0.013 0.051 ( t -statistic) ( - 5.96) (29.29) [# Positive] [2] [21] a See Table 1 for variable definitions. Firms are sorted into Low, Mod, and High categories as follows. For D Dispersion , in each calendar quarter firms with no change are assigned to group 5. Firms with negative (positive) D Dispersion are sorted evenly into groups 0–4 (6–10). Low; Mod; High = groups 0–2; groups 3–7; groups 8–10. For ST Revision , Low = ST Revision less than - 0.1%; High = ST Revision greater than 0.1%; Mod = ST Revision within the interval - 0.1% to 0.1% The stock price effects of changes in dispersion of investor beliefs 21 123
Image of page 21
The cost of capital hypothesis predicts that announcement period returns will be negatively associated with changes in the dispersion of investor beliefs. This implies that for every forecast error interval, returns should be higher as D Dispersion decreases. For 19 of the 21 forecast error intervals, we find that announcement period returns are higher for low D Dispersion than for high D Dispersion . We note that for one of the two intervals that produce results opposite from that predicted, the difference in returns is very close to zero. The other interval is the maximum forecast error interval, which includes a wide range of forecast errors and thus, it is difficult to know whether we have effectively controlled for the impact of the current earnings surprise. In Panel B of Table 4 , we formally test the difference in mean announcement period returns between the high and low D Dispersion categories across the 21 forecast error intervals. The difference of - 1.3% is easily significant at the 0.01 level and consistent with the cost of capital hypothesis . The importance of controlling for forecast revisions can be explicitly noted by considering the results for ST Revision in the final column of Table 4 . For all 21 forecast error intervals, announcement period returns increase as we go from low to high levels of forecast revisions for future earnings. The difference in means between the high and low categories is 5.1% and is easily significant. Thus, while
Image of page 22

Subscribe to view the full document.

Image of page 23
  • Fall '19

What students are saying

  • Left Quote Icon

    As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

    Student Picture

    Kiran Temple University Fox School of Business ‘17, Course Hero Intern

  • Left Quote Icon

    I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

    Student Picture

    Dana University of Pennsylvania ‘17, Course Hero Intern

  • Left Quote Icon

    The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

    Student Picture

    Jill Tulane University ‘16, Course Hero Intern

Ask Expert Tutors You can ask 0 bonus questions You can ask 0 questions (0 expire soon) You can ask 0 questions (will expire )
Answers in as fast as 15 minutes