Folks,
I am researching on the impact of a Corporate disclosure on future abnormal returns (in short term).
After reading few whitepapers, i realized one of the accepted way to measure the impact is removing the Normal Return portion from total return, the residual is Abnormal Return.
Theory sounds good to me, but my question is, how can i calculate the Abnormal return for a Stock? ( I am not well versed with understanding mathematical notation in whitepapers)
I have Closing Price, volume, Corporate news and its disclosure date.
Any guidance on that please?
You need a model that gives you the normal (expected) returns.
Usually this is the Capital Assets Pricing Model (CAPM) or related.
Cheers,
Dear Ankur Sancheti
Greetings,
Please look to the attached file about how to calculate the AR.
Best Regards
Hello Ankur,
The calculation of abnormal returns is simple. Firstly, you should understand that abnormal returns are the returns over the benchmark returns. In simple words, abnormal returns are stock returns minus the benchmark returns.
The benchmark could be any market index or portfolio against which you want to measure the performance of the stocks you are studying.
For calculation purposes, you needs to calculate the stock returns and the benchmark returns separately. Then subtract the benchmark returns from the stock returns.
Returns = (Closing price - opening price) / opening price
Abnormal returns = Stock returns - benchmark returns
The file attached by Faris would also be of help.
Please feel free to get back if you need further clarification.
Best,
You do not want to use complicated mathematical models and I can still say that you are lucky given the information you have to derive the abnormal returns. Abnormal returns occur following major corporate news or announcements such as increase in dividend payments or corporate mergers. Since, you have all the closing prices, volumes and corporate news and the disclosure dates, you simply have to study the behavior of the returns around the key announcement or disclosure dates. You have to filter your data by writing a simple program to access your data set. Once, you access the data set, examine all the returns around the announcement dates. You will be able to identify all the abnormal returns which are those returns around the announcement dates that are either positive or negative but with a very lhigh absolute value.
Hello,
I think you had very good remarks in previous comments.
I would like to add that according to my experience both Fama & French three-factor model, and Carhart four-factor model give slightly better estimates of expected returns than CAPM.
Also, you may need to cumulate the daily abnormal returns over several days (this period is called event window) in order to capture the entire effect of your event.
Best
You can calculate abnormal returns in several ways: CAPM, market model, etc. I suggest to read Brown and Warner (1980; 1985) for statistical test and Simon Benninga for practical application. I also suggest to read Fama, Fisher, Jensen and Roll (1969).
There are good contributions already. But one more point for those that want to employs the Fama and French model as their benchmark model for return. The values of alpha are the true values of the abnormal returns.
Lots of people use the market model (one single market portfolio, global, regional, or national), and the evidence indicate that it's quite robust. However, the markets may not be fully integrated, and that's why one needs to adjust for this issue. Thus, you can try a hybrid MM or a hybrid CAPM (combining global and regional/national indexes). See Bekaert and Harvey (1995) and more recently Andrieș et al. (2020).
- Badges
- Science topic
- Similar topics
- Filing