New answers tagged regression
Yes, the downside alpha is -0.0003. You can confirm that result if you include (1-D) explicitly as one more covariate, but then you'll have to request a model with no intercept.
Given that other corporate events are reasonably modelled through regression models (compare The Detection of Earnings Manipulation I would try for using an regression approach. I believe a more recent and related paper has been published but I don't seem to find it at this time. Edit: and now I did - Earnings Manipulation and Expected Returns That said, ...
Top 50 recent answers are included