Event Study - Impact from different periods


I was wondering if you could help me with something. I would really, really appreciate it. As it is some time since i did my econometrics course i hope this is not to trivial.

I'm currently doing an event study where i'm researching whether or not an earnings announcement (positive, negative or neutral compared to analyst estimates) impacts companies' stock prices. I want to compare impacts of earnings announcements before, under and after a financial crisis.

I will use these to calculate the effects:
Abnormal return = Actual return - Expected return
Cumulative abnormal return = Abnormal return over a certain time span
Average abnormal return = Averaged across all companies
Cumulative average abnormal return = Average across all companies over a certain time span

Test AAR:

So, my question is: With three time periods (pre financial crisis, financial crisis and post financial crisis), what's the best way to compare the returns from the different periods? The samples (pre, financial crisis and post) will be of unequal size.

Thanks in advance.