Multiple events within Company - Event Study

[Tony Yang]

Dear Stata users,

I am working on my master thesis Finance at the moment and I am trying to calculate the Cumulative Abnormal Return (CAR3) (i.e. [-1, +1] for each company ID. However, the companies researched has multiple announcement (i.e. event) at once or are announced the day after one has been announced. Does anyone know what command I should use? I looked up the Stata commands for a single event per company ID. This is noted as:
sort company_id date by company_id: gen datenum=_n by company_id: gen target=datenum if date==event_date egen td=min(target), by(company_id) drop target gen dif=datenum-td
by company_id: gen event_window=1 if dif>=-2 & dif<=2 egen count_event_obs=count(event_window), by(company_id) by company_id: gen estimation_window=1 if dif<-30 & dif>=-60 egen count_est_obs=count(estimation_window), by(company_id) replace event_window=0 if event_window==. replace estimation_window=0 if estimation_window==. Looking forward to hearing from you.

Kind regards,

Tony

[Phil Bromiley]

You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex. As you can see, what you've posted for code is very hard to read.

This is more of a finance substance question than a Stata question. Some finance studies drop observations with multiple announcements because they are hard to interpret. If you have multiple announcements, it only makes sense to try to estimate a single effect if the announcements are all about the kind of thing (e.g., mergers, income restatements). Otherwise, you're trying to estimate a parameter when the thing you're trying to estimate is not the same kind of thing across the events. You need to figure out the appropriate general sampling and estimation strategy. Once you have that, we can help you implement it.