Conducting event studies in Eviews

[mm44101]

I would like to conduct an event study for 500 companies. That implies calculating abnormal returns (AR) and cumulative abnormal returns (CAR) which are defined in the following manner:

AR(i,t)=R(i,t)- {a(i)+b(i)*R(m,t)}, where R(i,t) is the return of company i on day t, R(m,t) is the market return on day i, a(i) is the intercept of the simple OLS regression of R(i),R(m) during an estimation period before the event period and b(i) is the respective slope coeefficient. The estimation period is -300 to -41 days before the event. The event window comrises 21 days around the deal [-10,0,+10]. CAR(i,t) is defined as the sum of the 21 AR(i,t) during the event period.

My calculation requires 500 regressions with 260 days each and 500*21 calculations of the respective ARs and 500 calculations of the CARs. So far, I have calculated this in excel and importet the CARs into EViews for cross-sectional analysis. Now, I would like to switch to more sofisticated approaches to control for Heteroskedasticity in the residuals of the OLS regressions. Unfortunately I am not sure how to best implement my approach into Eviews.

How could I implement my data of 500*260+500*21 (stock returns) and 500*260+500*21 (market returns) and calculate the respective AR/CAR?

Thank you very much for any indication. Best,
Martin