I got a last "big" problem regarding my panel data. To give you a brief summary: I'm making an event study on nba games. Therefore, I createtd a panel to make certain tests in a regression to get the p-values and the return after each event. I did it with this code:
reg.ls r_ab_ c win_dummy_ product_playoff_win_
I wanted to look how a win in a game and a win in the playoffs affects the abnormal returns. Unfortunately only a few of my results were significant. Hence, my prof recommends me to use White as covariance method (estimate equation -> panel options -> coefficient cov. options). But there are three possibilities of White and I do not understand which one would be the best for my project. I had a look at the Eviews Guide II (p. 491-492 and 508-510), but to be honest it was not that helpful
Furthermore I tried to get the right code to implement it directly in my project. I tried it this way:
equation reg.ls.gmm(cxwhite) r_ab_ c win_dummy_ product_playoff_win_
But it doesn't work. Can anybody helps me out with this problem?
PS: Due to the fact that an event occurs almos every day in my project, I test only for one day after the event. Otherwise there would be too much overlapping of events.