Event study
method
Dr. Nguyen Thu Hien
This review is indebted to
MacKinlay, A. Craig, 1997, Event studies in
Economics and Finance, Journal of Economic
Literature, Vol XXXV (March 1997), pp. 13-39
Binder, John J., 1998, The Event Study
Methodology Since 1969, Review of
Quantitative Finance and Accounting, 11
(1998): 111-137
Corrado, Charles J, 2011, Event studies: A
methodology review, Journal of Accounting
and Finance, Vol. 51 (2011), 207234
Contents
Back ground
Method
Issue of model
I. An event study is
An event study is a statistical technique that
estimates the stock price impact of
occurrences such as mergers, earnings
announcements,
The basic notion is to disentangle the effects
of two types of information on stock prices:
information that is specific to the firm under
question (e.g., dividend announcement) and
information that is likely to affect stock prices
marketwide (e.g., change in interest rates).
Applications of event
studies
Earliest event study: Dolley (1933)
Ball and Brown (1968) and Fama et al.
(1969) planted seeds of financial research
that continue to flourish decades later
No one really knows how many event
studies have been published.
Kothari and Warner (2005) report that over
the period 19742000, five major finance
journals published 565 articles containing
event study results
Applications of event
studies
In practice, event studies have been used for
two major reasons:
1) to test the null hypothesis that the market
efficiently incorporates information (see Fama
(1991) for a summary of this evidence)
2) under the maintained hypothesis of market
efficiency, at least with respect to publicly
available information, to examine the impact
of some event on the wealth of the firm's
security holders.
Outline of an event study
1.
2.
3.
4.
5.
6.
7.
Event definition
Selection criteria
Normal and abnormal returns
Estimating procedure
Testing procedure
Empirical results
Interpretation and conclusions
II. Models for measuring normal
returns
Constant-mean-return model
Market model
Other statistical model
Economic models
Constant-mean return
model
Assume mean return is unchanged, with
constant variance
Simplest model, often yields resutls similar
to more sophisticated models (Brown,
Warner 1980, 1985).
May be because: variance normally not
reduce much using more
Market model
Assume joint normality of asset returns
(stock market and the company stock)
By removing the portion of the return that is
related to variation in the markets return,
the variance of abnormal return is reduced.
increase ability to detect event effects.
Higher R2, greater is the variance reduction
of the abnormal return, larger is ability to
detect abnormal return.
Measuring and analyzing abnormal
return
Estimation of the market model
Statistical properties of abnormal
returns
Aggregation of abnormal returns
Economic models
CAPM: used porpularly in 1970s.
Restrictions imposed on stock returns
(using market risk premium) is
questionable (Fama, French 1996).
Possibly, results the studies may be
sensitive to the specific CAPM
restrictions. Use of CAPM in event
studies has almost ceased.
Economic models
APT (Arbitrage Pricing Theory): with
the APT, the most important factor
behaves like a market factor and
additional factors add relatively little
explanatory power.
Gains from using APT versus the
market model are small.
III. Event window and
Measuring abnormal returns
t= 0: event date
Estimation of the market
model
Under general conditions ordinary least
squares estimation (OLS) is a
consistent estimation procedure for the
market model parameters.
Abnormal return estimation using
market model:
AR is disturbance term of the market
model calculated on an out of sample
basis.
Estimation of the market
model
Variance of abnormal return:
First term is variance of disturbance in
market model (equation (3)
Second term is additional variance due
to sampling error in
and
.
When estimation window is long (large
L1), second term approaches zero.
Aggregation of abnormal
returns
To draw overall inferences for the
event, aggregation of abnormal return
observations is helpful.
(1) Aggregate through time and (2)
across securities
Aggregating thru time can be done by
calculating CAR:
Aggregation of abnormal
returns
Aggregating across security can be
done by calculating average AR:
Aggregation of abnormal
returns
Aggregating both thru time and
across securities:
Hypothesis testing: H0: the event has
no information impact:
Further issues
Role of sampling interval
Inferences with event-date
uncertainty
Possible biases
Assignments
Group assignments:
1/ read and present 2 papers
2/ test effect of M&A announcement
on VN stock market