Empirical Methods in Finance 323063
Event Studies
Event studies: test of change in stock (or bond) prices around specific ‘events’
(corporate earnings announcements, merger announcements, macro-economic
news, etc.), announcement is often more important than actual event, major
reasons to use event studies:
1. Examine impact of event on wealth of firm’s security holders.
2. Test hypothesis that market efficiently incorporates new information (EMH).
Efficient Market Hypothesis: prices always fully reflect available and relevant
information, prices will change only when new information arrives, information
(therefore, price changes) cannot be predicted ahead of time by definition.
Efficiency refers to two aspects of price adjustment: speed and quality (direction
and magnitude). Primary of capital market is allocation of ownership of economy’s
capital stock, assets prices should provide accurate signals for resource allocation.
● Weak form: prices reflect all information contained in record of past prices technical
analysis. ⇒ tests for return predictability.
○ Prices technical analysis → time series.
○ Fundamental variables → cross section.
● Semi-strong form: prices reflect not only past information, but all other published
information, adjustments are:
○ Direct → event studies;
○ Complete → event studies.
● Strong form: prices reflect not only just public information, but any information that
might be relevant ⇒ tests for insider/private information.
Idea: in case of semi-strong market efficiency, there should be stock reaction on day
announcement is realized (unless the two correspond).
Identify three steps:
1. Identify event of interest and timing of the event. Timing: notice difference
between calendar time and event time: t=0 is time of event, typically event study
uses many (N) observations on similar events that can be far apart on calendar, but
event time is used.
2. Specify a “benchmark” model for normal stock return behaviour (normal market
fluctuations).
3. Calculate and analyse abnormal returns around the event date.
Two relevant announcements on the same day: merger announcement of oil firm +
oil prices announcement. Usually don’t use this date in event studies.
Estimation and Event Window
,
Event Studies
Event studies: test of change in stock (or bond) prices around specific ‘events’
(corporate earnings announcements, merger announcements, macro-economic
news, etc.), announcement is often more important than actual event, major
reasons to use event studies:
1. Examine impact of event on wealth of firm’s security holders.
2. Test hypothesis that market efficiently incorporates new information (EMH).
Efficient Market Hypothesis: prices always fully reflect available and relevant
information, prices will change only when new information arrives, information
(therefore, price changes) cannot be predicted ahead of time by definition.
Efficiency refers to two aspects of price adjustment: speed and quality (direction
and magnitude). Primary of capital market is allocation of ownership of economy’s
capital stock, assets prices should provide accurate signals for resource allocation.
● Weak form: prices reflect all information contained in record of past prices technical
analysis. ⇒ tests for return predictability.
○ Prices technical analysis → time series.
○ Fundamental variables → cross section.
● Semi-strong form: prices reflect not only past information, but all other published
information, adjustments are:
○ Direct → event studies;
○ Complete → event studies.
● Strong form: prices reflect not only just public information, but any information that
might be relevant ⇒ tests for insider/private information.
Idea: in case of semi-strong market efficiency, there should be stock reaction on day
announcement is realized (unless the two correspond).
Identify three steps:
1. Identify event of interest and timing of the event. Timing: notice difference
between calendar time and event time: t=0 is time of event, typically event study
uses many (N) observations on similar events that can be far apart on calendar, but
event time is used.
2. Specify a “benchmark” model for normal stock return behaviour (normal market
fluctuations).
3. Calculate and analyse abnormal returns around the event date.
Two relevant announcements on the same day: merger announcement of oil firm +
oil prices announcement. Usually don’t use this date in event studies.
Estimation and Event Window
,
