Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • (unusually?) large coefficient

    Hello!

    I'm running a probit regression. The DV, continue, is a dummy variable indicating whether the firm would choose to continue doing sth (did this thing in year t-1, and continue doing this in year t); the IV, Δinstitutional, is the change of the percentage of a firm's shares held by institutional investors from year t-1 to t. I have some control variables also measured as change variables. The purpose of this regression is to investigate how the change in a firm's shares held by institutional investors from year t-1 to t affects the firm's decision to continue doing sth. However, I obtain a very large coefficient of Δinstitutional, I wonder what could probably result in this large coefficient, is it reasonable? Is that because the DV is not a change variable whereas the IV and controls are change variables, so they are not comparable?
    Code:
                       |               Robust
              continue |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
    -------------------+----------------------------------------------------------------
         institutional |
                   D1. |    8.83314   12.78101     0.69   0.489    -16.21719    33.88346
    Thanks a lot for any help!
Working...
X