Event Study in Stata

[Milica Matovic]

Dear all,

I am very new to Stata and I hope someone can help me as I've looked everywhere and couldnt find anything to answer my question.

I want to import my data and set it up in a format of event study. I am looking at company additions to the S&P 500 and their stock returns before and after the event.
I also have two events of interest, announcement day (when S&P announces the change) and effective change day (when the change actually takes place).
I am looking at the abnormal returns and volumes.

Could anyone please help?
Grateful in advance!

[Phil Bromiley]

You'll increase your chances of a helpful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex.

If you searched the Forum or Googled the issue, you would find that this has been discussed many times. It is appropriate to do both before posting a question.

[David Benson]

Hi Milica, as long as you have the CUSIP or Permno for the stock, the announcement day and the effective change day, your request is really easy to run within CRSP / EVENTUS (as long as you or your university have access to WRDS. Eventus is a WRDS program that uses CRSP stock market data to run event studies (it lets you run Cumulative Abnormal Return and Trading Volume event studies). It also lets you export that data so you can then do hypothesis testing (i.e. is the effect more pronounced when announced rather than implemented? Is it larger when the founder is the CEO?)

BTW, I suspect you will find a larger effect on the announcement day (when S&P announces the change) than on the actual change day (when the change actually takes place). McWilliams, Turk, and Zardkoohi (1993), in a paper examining the effects of Supreme Court decisions on pending mergers, found that most of the market reaction occurred when the case was argued, rather than when the Supreme Court's decision was announced.