Event-study, Kothari & Warner + MacKinlay

Question 1

What does event studies examine?

Answer 1

Event studies examine the behavior of firms’ stock prices around corporate events

Question 2

What is the relationship between event studies and testing market efficiency?

Answer 2

Event studies serve an important purpose in capital market research as a way of testing market efficiency. Systematically nonzero abnormal security returns that persist after a particular type of corporate event are inconsistent with market efficiency. Ac- cordingly, event studies focusing on long-horizons following an event can provide key evidence on market efficiency

Question 3

What does an event study seek to establish?

Answer 3

An event study seeks to establish whether the cross-sectional distribution of returns at the time of an event is abnormal (i.e., systematically different from predicted).

Question 4

What is the typical specific null hypothesis to be tested in event study?

Answer 4

Typically, the specific null hypothesis to be tested is whether the mean abnormal return (sometimes referred to as the average residual, AR) at time t is equal to zero.

Question 5

What is the cumulative average residual method (CAR)?

Answer 5

The cumulative average residual method (CAR) uses as the abnormal performance measure the sum of each month’s average abnormal performance. (abnormal = the difference between actual and expected performance)

Question 6

What is an Type I error in event study?

Answer 6

A Type I error occurs when the null hypothesis is falsely rejected

Question 7

What is an Type II error in event study?

Answer 7

A Type II error occurs when the null is falsely accepted

Question 8

What does it mean that event study tests are joint tests?

Answer 8

event study tests are joint tests of whether abnormal returns are zero and of whether the assumed model of ex- pected returns (i.e., the CAPM, market model, etc.) is correct. Moreover, an additional set of assumptions concerning the statistical properties of the abnormal return measures must also be correct.

Question 9

What does abnormal return mean?

Answer 9

These are the differences between actual returns (what actually happened) and expected returns (what should have happened based on a model).
Example: If a stock is expected to rise by 2% but rises by 5%, the abnormal return is +3%.

Question 10

What are the two joint tests in event study?

Answer 10

Event studies simultaneously test two things:

1. Are the abnormal returns actually zero?
   This checks if the event had any effect.
   The assumptions about the statistical behavior of abnormal returns must hold true.
2. Is the expected return model correct?
   This means the method used to calculate “expected returns” (e.g., CAPM or market model) needs to work properly. If the model is wrong, the abnormal returns might be misleading.

The model used to calculate expected returns (e.g., CAPM) must be correct.

Question 11

What model can you use pre-event to predict expected market returns?

Answer 11

You use a model (like CAPM) to predict expected returns based on market trends.

Question 12

What is a t-test used for in event study?

Answer 12

A t-test is used to determine if the average abnormal return (across several securities or over time) is significantly different from zero.

Question 13

What are the assumptions for the T-test?

Answer 13

Normal distribution and independence.

Normal Distribution: The abnormal returns are assumed to follow a bell-shaped curve (normal distribution).

Independence:
Across time: Abnormal returns for one day shouldn’t depend on those of another day.

Across securities: Abnormal returns for one stock shouldn’t depend on those of another stock.

Question 14

What can be the problem with small samples in event studies?

Answer 14

When the number of observations (stocks or days) is small, you can’t rely on the central limit theorem (which usually helps data behave normally when sample sizes are large).
In small samples, these assumptions need to be checked carefully because violations can make the t-test results unreliable.

Question 15

What is the Brown-Warner simulation?

Answer 15

its a method used in event study research to test the reliability of event study methods. The idea is to use simulations with actual financial return data to understand how well event study tests work.

Instead of analyzing real-world events, the method uses randomly selected securities (stocks) and random event dates.
If everything is working correctly, these randomly created “events” should show no abnormal performance (because they are random and unrelated to any real event).

Question 16

Why do long-horizon event study tests often fail to detect abnormal performance?

Answer 16

because returns become noisier over longer periods.
True effects of an event might get “hidden” by other unrelated market factors.

Short-horizon tests are more robust, meaning they are less affected by the choice of the benchmark model (e.g., CAPM) or assumptions about how returns behave.

Question 17

What are Cross-Sectional tests in event studies?

Answer 17

These tests look at how abnormal returns (unexpected stock price changes) differ between firms.
They analyze whether firm-specific factors (like size, industry, or how closely analysts follow the firm) explain the variation in abnormal returns.

Question 18

Which are the two main methods for assessing and calibrating post-event risk-adjusted performance for long-horizon event studies?

Answer 18

1. Characteristic-based matching approach
2. Calendar-time approach (Jensens Alpha)

Question 19

What is the Characteristic-based matching approach?

Answer 19

Compare the performance of the event firm’s stock to that of similar firms (matched based on characteristics like size, industry, and book-to-market ratio).
Example: If a small tech company announces a merger, you compare its post-event returns to other small tech companies that didn’t have a merger.
Goal: Identify the abnormal performance by seeing how the event firm differs from its peers.

Question 20

What is the Calendar-time approach (Jensens Alpha)?

Answer 20

Group event firms into a portfolio and calculate the portfolio’s alpha (a measure of abnormal performance) using a risk-adjusted model like CAPM or Fama-French models.
Example: Combine all merger firms into one portfolio and analyze the portfolio’s risk-adjusted performance over time.
Goal: Assess whether the portfolio consistently outperforms or underperforms the market after accounting for risk.

Question 21

What is the BHAR approach?

Answer 21

Characteristic-Based Matching Approach, often referred to as Buy-and-Hold Abnormal Returns (BHAR)

BHAR measures the long-term abnormal returns of firms that experience a particular event (like a merger or stock buyback) compared to firms that don’t.

Question 22

Long-horizon buy-and-hold returns, even after adjusting for the performance of a matched firm (or portfolio), tend to be WHAT skewed?

Answer 22

right skewed

Question 23

What is an event study?

Answer 23

Using financial market data, an event study measures the impact of a specific event on the value of a firm. The usefulness of such a study comes from the fact that, given rationality in the marketplace, the effects of an event will be reflected immediately in security prices.

Question 24

What is the null hypothesis in event studies?

Answer 24

that the event has no impact on the distribution of returns

Question 25

What is the event window?

Answer 25

the period over which the security prices of the firms involved in the event will be examined. often, the period of interest is often expanded to multiple days, including at least the day of the announcement and the day after the announcement. This captures the price effects on announcements which occur after the stock market closes on the announcement day.

Question 26

What is the abnormal return defined as?

Answer 26

the actual ex post return of the security over the event window - the normal return of the firm over the event window

Question 27

What is the normal return defined as?

Answer 27

the expected return without conditioning on the event taking place

Question 28

Which are the two common choices for modeling the normal return?

Answer 28

the constant mean return model (assumes that the mean return of a given security is constant through time)

and the market model (assumes a stable linear relation between the market return and the security return): Uses a stock’s relationship with the market index to predict normal returns. Most commonly used due to its simplicity and effectiveness in reducing noise by accounting for market-wide effects.

Question 29

How can you define the estimation window?

Answer 29

using the period prior to the event window. generally, the event period itself is not included in the estimation period to prevent the event from influencing the normal performance model parameter estimates

Question 30

What is another statistical model to model normal return, besides the market model and the constant mean return?

Answer 30

factor model. they are motivated by the benefits of reducing the variance of the abnormal return by explaining more of the variation in the normal return. the market model is an example of factor model

Question 31

Do event study rely on efficient markets?

Answer 31

Yes, event studies rely on the assumption of efficient markets, where prices fully reflect all available information.

Question 32

What is CAR?

Answer 32

The sum of abnormal returns over the event window to assess the total impact of an event