"day-of-the-week" stock price effect - assignment problemWatch
I would much appreciate if someone was able to help me with this terrible econometric assignment.
I am supposed to analyse if abnormal returns on some chosen industry index on days falling 'around weekend' are significant.
My idea: calculate abnormal returns for Fridays and Mondays.
To calculate the expected returns, I am planning to use CAPM model.
Question number 1: how to calculate beta?
That is, should I take to the "Friday results" beta calculations only market returns that occurred during those Fridays?
Or should I take under consideration all the market returns, even those falling on Tuesdays, Wednesdays, etc.? - this way, I will get more entries (data) on market returns, than I would have on Fridays returns...
Question number 2: (silly one) I am supposed to get just one beta, right? Not a separate one for each Friday, for instance?
Next, I understand, that I have to use regression, to check whether the abnormal results on Friday and Monday are significant. My plan is that I will in regression use (here, as an example for 'Fridays'):
- as 'Y' - abnormal returns for Fridays
- as 'X' - all the observed market returns that took place on Fridays
Question number 3: is this how I will be able to test the significance of abnormal returns for Fridays or I got it completely wrong?
I would be extremely grateful for any information provided!
BTW, spoiler alert - this is my first regression and CAPM use in the whole life time