Transcript Efficient MarketsII

Class Business


Homework
Upcoming Midterm
– Review Session
•
Wed (5/18) 5 – 6 pm 270 TNRB
Predictability of Stock Returns

Momentum and reversal might happen due to “fads” and
“bubbles”:
–
Stocks over-react to news initially and correct to go back to
fundamental value.
–
‘Overconfidence’, ‘herding’, and ‘biased self-attribution’
Adjusted Cumulative Returns of Takeover
Target Companies
Source: Keown and Pinkerton (1981)
Returns following Earnings
Announcements
Source: Patell and Wolfson (1984)
EMH and Competition

Stock prices fully and accurately reflect available information is a static
concept

Once information becomes available, we expect market participants
analyze it quickly ( ‘time’ now enters argument)

Competition assures prices quickly reflect information

Most interesting questions deal with how quickly prices adjust to new
information or sets of information

There is probably an ‘optimal’ amount of inefficiency
– Marginal cost of finding the inefficiency vs. marginal benefit
Blind Monkeys Throwing Darts

Some efficient market supporters suggest that it does not
matter how you choose stocks in efficient markets.

You can ask some blind monkeys to throw darts on the Wall
Street Journal to select stocks.
Blind Monkeys Throwing Darts

However, rational security analysis is still useful:
–
Monkeys will probably not pick a well-diversified
portfolio with an desired level of risk.
– Monkeys do not know the tax considerations of
stock choice.
–
Monkeys do not take your specific
circumstances into account (job, age, location).
More...Are Markets Efficient?

Magnitude Issue
–
Some price anomalies are relatively minor, because transactions
costs exceed the profits
–
Example:
•
•
•
•


Suppose you have $10,000 invested
Suppose research will yield a guaranteed increase in return of 0.1%
over the next year with no increase in risk.
You get $10!
Now suppose you have $10 billion invested . .
Selection Bias Issue
–
Investors will not report the successful strategies.
Lucky Event Issue
–
If you have enough people doing something, then some people
will always be lucky.
Market Anomalies



Small-Firm Effect
–
Stocks of small firms have earned abnormal
returns (particularly in January).
Book-to-Market Effect
–
Stocks with high book-to-market ratios (value
stocks) tend to outperform stocks with low bookto-market ratios (growth stocks).
P/E Effect
–
Stocks with low price-earnings ratio (value
stocks) tend to outperform stocks with high
price-earnings ratio (growth stocks).
Small-Firm Effect
Source: Ibbotson Associates (2000)
Value versus Growth Stocks
Source: Fama and French (1992)
Interpretation of
Market Anomalies

Market Anomalies are due to:
–
Risk Premia
•
–
Irrational Behavior
•
–
Are we accounting for all the appropriate risk factors,
such as in an multifactor framework? (more factors than
just market portfolio)
Investors prefer to purchase large and growth stocks and
neglect small and value stocks.
Data Mining
•
By chance, some criteria will appear to predict returns.
S&P 500
The S&P 500
1983-1993
Year
Source for all the S&P 500 data mining graphs is: David Leinweber’s “Data-Snooping Biases in Tests of Financial Asset Pricing Models.”
Overfitting the S&P 500
S&P 500
Butter in Bangladesh
R2=.75
Year
Overfitting the S&P 500
S&P 500
Butter Production in Bangladesh and the United States
R2=.95
Year
Overfitting the S&P 500
S&P 500
Butter Production in Bangladesh and the United States
United States Cheese Production
Sheep Population in Bangladesh and the United States
R2=.99
Year
Data Mining



Technical Analysis
Motley Fool
– Foolish Four
– Dow 30
Need good economics to validate a trading strategy,
not just an empirical relationship in historical data
EMH – How do we define risk?




It is impossible to determine how efficient markets are by looking
at returns.
“Anomalies” depend on which model you believe is correct for
expected returns.
If you uncover an anomaly, EMH supporters can always claim
you just don’t have the right model or measure of risk.
A better way to test for efficiency is to look at performance of
active managers.
Mutual Fund Performance:
Alpha

Actively managed mutual funds earn alphas that are slightly
negative, but not statistically different from zero.

Some evidence that a few rare managers can earn abnormal
returns. (e.g. Warren Buffett)

Evidence indicates that, unless you have confidence (information)
in picking a manager, you’re better off in a passive index than
chasing the hottest fund manager.