Equilibrium in Financial Markets - Tests of Market Efficiency and Annomalies Flashcards
What is an anomaly?
Is a cross-sectional and time series pattern in security returns not predicted by a particular asset pricing model - leads to a persistent and predictable realization of abnormal returns
What constitutes an abnormal return?
The difference between the actual returns of a security/portfolio and the expected return
What is an excess return?
Difference between the actual return of a security and the riskless rate
What are the 4 tests of market efficiency?
1 - Time series properties of price changes
2 - Market reaction to information events
3 - Cross-sectional and seasonal anomalies
4 - Evidence on insiders, analysts and investment managers
For the test of Time Series Properties of Price Changes, what is the general approach?
Assess if stock prices are predictable -> change in periods of hours or days and if provide information for future prices
For the test of Time Series Properties of Price Changes, what are the main conclusions of the serial correlation?
-correlation between prices changes in consecutive periods
-serial correlation of zero implies investors can learn about future prices from past ones
- positive serial correlation viewed as evidence of price momentum - positive returns more likely to follow positive returns
-negative serial correlation viewed as evidence of price reversals
- there can be serial correlation without any opportunity to earn abnormal returns
What is a filter rule?
an investor buys an investment if the price rises x% from the previous low and holds the investment until the price drops x% from a previous high
For the test of Time Series Properties of Price Changes, what concerns the technical analysis and filter rules?
Investigating patterns based on historical trading data to identify the future movements of stocks or the market
Resistence and support levels should not easily be crossed
Trading strategies are serially correlated and there is price momentum
For the test of Time Series Properties of Price Changes, how do run tests work?
They are non parametric tests based upon a count of the number of runs -> the actual number is compared with the expected, assuming price changes are random. If the actual number is higher, there is evidence of a negative correlation in price changes
For the test of Time Series Properties of Price Changes, and the study of price changes for different periods, how does it go in the short term?
periods up to a month - there is evidence of reversals in stock prices - if returns have been positive recently, in the following months returns are more than likely going to be negative
For the test of Time Series Properties of Price Changes, and the study of price changes for different periods, how does it go in the short/medium term?
Evidence that prices tend to move in the same direction over horizons of months and underreact to information - momentum effect - negatively related to the market capitalization
Where is the momentum effect more pronounced?
Among low-volume winners and high volume losers
What are some of the possible reasons of momentum effect?
-Slow information diffusion - markets don’t absorve information right away
-Investors may be slow to react to new information
What may cause the slow reaction from investors which may be one of the causes for the momentum effect?
1-overconfidence - investors will think their information is very valuable and tend to disregard public information
2-conservatism bias - don’t revise their expectations as frequent as thei should
3-interaction of two classes of agents who process information in different ways - news watchers vs momentum traders (buy past winners and sell past losers)
4-disposition effect - tendency to stick to loosers and sell winning stocks. This happens because if they are overconfident they don’t want to admit they made a mistake in the past
5-anchoring on past prices - and don’t update their views as much as they should
A study concluded that momentum profits depend on the state of the market - where do they tend to exist (when is it more favorable)?
When markets are in a tendency to increase - bull markets; also up markets and high volume stocks
Why is price momentum related to the uncertainty in information?
the more uncertain the information, the bigger the momentum effect is, because prices will take longer to absorve that information (related to the limitations in interpreting information)
A study found that national cultures are important to explain the momentum effect, why?
- individualism is positively associated with momentum profits - if investors are less individualistic, they’ll put less weight on information they come up with, and more on the consensus of their peers
-the higher the level of overconfidence in a country, the more impact the momentum effect will have
When do momentum crashes occur?
When there has been a market decline and so market volatility is high
What does the effect of investor sentiment has to do with momentum effect?
news that contradict investor’s sentiment causes cognitive dissonance, slowing the diffusion of signals that oppose the direction of sentiment. This causes underreaction to negative info under optimism and to positive pieces of info under pessimism
regarding momentum effect, past winners can be classified into which two different categories?
-unconstrained past-winner stocks - high past returns reflect positive fundamental value stocks - underreaction produces momentum
-constrained winners - high past returns reflect a sentiment/disagreement shock - competition among optimistic investors combined with short-sale restrictions lead to a rise in the price and when optimism wanes, prices fall leading to negative returns
Regarding the study of price changes in the long term what are the main conclusions taken?
We can see that there is evidence of price overreaction and mean reversion - past winners will underperform past loosers - they got overvalued and will return to the mean
The presence of momentum in the short/medium term and mean reversion in the long term are contradictory. How can we reconcile these two?
1) Behavioral models - conservatism and representativeness bias; overconfidence and self-attribution bias; existence of feedback traders and news watchers; and aversion effect
2)Momentum Life Cycle Hypothesis - low volume stocks have a very low search, so they are cheap. At a certain point they will call the atention of investors, getting a higher demand and will enter the stage of low volume winners. The price will increase, atracting more investors, increasing the volume of trading, entering the stage of high volume stocks. At a certain point, these stocks will be overvalued and will disapoint -> price will go down -> high volume loosers - a lot of investors trading those stocks, but with low return, and will end up again as low volume loosers
What are event studies?
A study on the market reaction to public information and the existence of abnormal returns after the disclosure of information - if the markets are efficient in the semistrong form - the price reaction must be instantaneous and not biased
What are the main evidence from earnings announcement studies?
-evidence of underreavtion to news -> positive excess returns after positive announcements and negative excess returns around negative announcements PEAD
-unexpected high returns outperform unexpectedly poor earnings
-evidences of insider trading
-timed earnings reports - time to process the information
-positive relation between PEAD and the level of individualism in a country
What are the main critiques of event studies?
- long term event studies suffer from the joint hypothesis problem
- short term event studies are only able to detect special forms of inefficiency
- we can get false negatives
- event study methods exhibit a bias toward detecting effects
-event studies try to capture the linear impact of information when the market may be driven by the nonlinear processes
What regards the joint hypothesis problem?
We don’t know what model to use, if it’s the correct one
Emprirical evidence shows that there is a continuation of prices in the short term. They behave consistently with the underreaction to information in the short term. This underreaction manifests in what two ways?
1) underreaction to private information - price momentum - positive correlation between individual returns and the market as a whole
2) underreaction to public news events - abnormal returns in case of PEAD