Reply With QuoteIt sountds to me like you are making predictions in time in which case regression will have problems (simple regression) if you have autocorrelation or there is a trend in your data (or seasonality).
Dear all,
at the moment I'm doing a research on Merger and Alliances. I'm doing an Event Study, so checking the cumulative abnormal returns(CAR) of the stocks of the given firms. My first hypothesis I'm testing is "Acquiring firms performing related acquisitions will have higher abnormal returns than acquiring firms performing unrelated acquisitions." Relatedness is a simple dummy with a value of 0 or 1. To test this I use an Independent sample t-test.
My questions now concerns my second hypothesis "The size of the acquiring firm will not influence the abnormal returns in a related acquisition." Is is correct to purely apply a regression with the CAR as independent variable, and relatedness and firm size as dependent variables?
Thank you very much for you Help!
It sountds to me like you are making predictions in time in which case regression will have problems (simple regression) if you have autocorrelation or there is a trend in your data (or seasonality).
"Facts are stubborn things, but statistics are more pliable." Mark Twain
Thank you noetsi. So if I check for autocorrelation and/or seasonality and they are of no matte, then it is okay to use the regression? Or how do I deal with this?
Posting Permissions
|
|