Behavioral Finance Exam 1
What are the Fama-French Factors in the three factor model? What additional factors are in the five factor model?
(1) MKT - market risk premium (2) SMB (small minus big) - the outperformance of small-cap companies relative to large-cap companies (3) HML (high minus low) - the outperformance of value stocks (high book-to-market value companies) vs growth stocks (low book-to-market value companies) (4) RMW (robust minus weak) - the outperformance of stocks with robust (high and steady) profitability vs those with weak (low) profitability (5) CMA (conservative minus aggressive) - the difference in performance between stocks with a high and low reinvestment rate market risk, cap, value, profitability, reinvestment
Traditional finance acknowledges that individuals have behavioral biases but believes that rational models predict behavior better based on what 4 factors? What is the response to these arguments by people who believe in behavioral finance?
- irrationalities "cancel out" - people act rationally with big $$$ at stake - important players in markets are likely to be rational - smart money wins in the end The response of people who believe in behavioral finance are that there are limits to arbitrage and there is not enough rational money in the market. And that people have behavioral biases that have a correlation to prices in the stock market.
If markets are efficient, what value should the correlation coefficient take between stock returns for two non-overlapping periods?
0
Give five reasons why indices are constructed?
1. Benchmark for measuring performance 2. Used as the basis for creating financial products such as derivatives 3. Market analysis such as studying market trends and tracking historical performance 4. Indices are used to allocate assets: stocks vs bonds, US versus non-US, value vs growth, etc 5. Risk management such as assessing risk and volatility of different market segments
React to the following statements: A. If stock prices follow a random walk, then capital markets are no different than a casino. B. A large part of a company's future cash flows are predictable, therefore its stock price cannot possibly follow a random walk. C. If markets are efficient, you might as well select your portfolio by throwing darts at the stock listing in the Wall Street Journal (paper edition...)
A. false - capital markets follow a random walk and incorporate information B. false C. true
What is the difference between "active" management and "passive" management? What does "active" management require beyond investor irrationality?
Active means you assume the market is inefficient or wrong in some way and you trade based on that. Passive means you believe the market is efficient and you trade based on that. It does not refer to the amount or frequency of trades. Active management also requires that markets are correlated and predictable.
Why might agency traders be less subject to the biases of overconfidence than proprietary traders?
Agency traders don't make the decision to buy or sell
What effect does professional trading have on the disposition effect?
Agency traders often don't make the decisions to buy or sell, so the effect is minimized
How is the Russell 3000 index constructed? How does it differ from the S&P 500?
Comprised of 3,000 publicly held US companies that represents about 98% of the investable US equity market. These companies are ranked by market cap (share price x shares outstanding) and weighted based on this - so higher market cap = higher weight. The S&P 500 is comprised of 500 of the most widely traded stocks in the US and represents about 70% of the total value of US stock markets. It is also less objectively decided and how it's decided is not made public. The Russell 3000 is more frequently and carefully adjusted.
Differentiate between gambling fallacy and clustering illusion. How do these relate to the representativeness heuristic?
Gambling fallacy is believing that a repeated occurrence of something makes it less likely to happen in the future. Clustering illusion is perceiving patters or clusters in random data, and assuming that pattern is not truly random and is bound to continue. The representativeness heuristic is when we base the probability of something on a prototype or stereotype that we already have in mind.
How does our motivations place obstacles in the path of risk perception?
Our motivations may encourage us to seek out information that supports that (confirmation bias) and can cause us to avoid information that signifies risk or goes against our motivations
How does prospect theory help explain the "equity premium" puzzle?
People feel losses more than gains and need compensation for it
What is "bounded rationality"?
People have goals they want to achieve - and are satisfied when they got those as opposed to trying to maximize.
Give examples of how people over-estimate the probability of rare events and underestimate the probability of frequent events. What factors determine by how much they do so?
People overestimate the probability of nuclear events. People underestimate the probability of crashing on a bike. The perceived level of dread determines the over or underestimation.
How does overconfidence relate to trading and what affect does it have on a portfolio's return?
People trade too much, incur too many transaction costs, lowers portfolio returns
What predictions does prospect theory make for losses and gains?
People value avoiding losses far more than they want gains.
Describe the effect of cognitive dissonance on the propensity to save
People who are older still think they are young - so they feel they don't yet need to save
Define the gambler's fallacy and distinguish it from the regression fallacy
Regression fallacy - when something unusual happens, it will probably go back to normal next time. Gambler's fallacy - the belief that a random event is influenced by past outcomes
What is data snooping?
Searching through and manipulating large amounts of data until you find something of statistical significance.
Give a general definition of framing and discuss the reasons for framing?
Set of heuristics to interpret the world around us. We use it to find a way to condense, understand and interpret all of the information around us.
What is skewness and kurtosis?
Skewness is a measure that tells you if the values in a data set are stretched more to one side of a distribution. Kurtosis measures the concentration of extreme measures in the tails of a data set - whether there are a lot of extreme values.
Define the concept of a certainty equivalent?
Someone would rather accept a guaranteed return that was lower, than take a chance on a higher, but uncertain, return. The risk premium is the difference between the expected value and the certainty equivalent.
What is the definition of "SUE"? How might we take advantage of this anomaly?
Standardized unexpected earnings - we take (what's announced - what is expected) and divide it by something (typically price). Very high SUE stocks generally overperform after the earnings announcement and ones with low SUE generally underperform
Describe the two steps that investors use to evaluate investments if they are following prospect theory.
Step 1 - frame the outcomes in terms of gains or losses Step 2 - the S-shaped value function, which is risk-seeking in the loss portion and risk-averse in the gain portion
Define System 1 and System 2. What is it useful to make the distinction between these two ways of thinking?
System 1 - thinking that happens automatically, intuitively and with little effort. Driven by instinct and experiences. System 2 - slower and requires more effort. Conscious and logical.
What assumptions does expected utility maximization make that are questionable?
That people know their utility functions. That people know the outcomes of the risks they take. That people know the probability of the outcomes. That we simply maximize our utility - people spend more time figuring out what their options are rather than determining maximum utility
Briefly describe how the following create obstacles to better risk perceptions: a. Hyperbolic discounting b. Finite cognitive capacity c. Endowment effect d. Memory e. Cognitive dissonance f. Salience
a. we value short-term reward over long-term reward, even if the long-term reward is clearly better b. we can only do so much of something c. status quo bias d. we remember the things that hurt us the most but that may not be what's most probable e. we think we are different than what we actually are and that changes our perception f. we overstate probability based on how easy something is to recall
What is the certainty equivalent?
The guaranteed amount of money that a person would consider as having the same amount of desirability as a risky asset
What are the assumptions of the market model?
Error term is 0, relationship is linear, intercept is the risk free rate Statistical assumptions - the error term is 0, meaning the omitted variables do not affect the market. Another assumption is that there is a linear relationship rather than a non-linear one. Another assumption is that the intercept is the risk free rate.
What is your response to a co-worker who says "Highly variable stock prices suggest that the market is inefficient because it does not know how to price stocks?"
Not true bc highly variable stocks could have a beta that allows you to price them. And market pricing does not mean zero volatility, it means pricing relative to systematic risk
Label the following statements as true or false if the efficient markets hypothesis is true: a. The EMH implies that future events can be accurately predicted. b. The EMH implies that the average rate of return will be greater than zero. c. The EMH implies that patterns in past prices will predict future prices. d. Earnings announcements will only effect the price at the time of the announcement and will have no effect on future prices. e. You can make superior returns by buying a stock after a 10% rise and selling a stock after a 10% fall
A - false B - false C - false D - true E- false if the market's efficient
If a stock has a βSMB= -1.2 and a βHML= -0.75 what type of stock is it?
A large growth company
What is the definition of the word "statistic?" What problem does a statistic attempt to solve.
A statistic is a function of the sample data - and a number that represents a piece of information. Statistics attempt to find relationships and help to better understand unknown (random) variables
What is an investment philosophy?
A theory of how the market operates and a set of strategies that defines that theory. Largest categorization is active vs passive. Can be further broken down into strategies such as growth, value, arbitrage, earnings management, etc.
Are the following true or false? A. Stocks with a beta of zero offer an expected return of zero. B. The CAPM implies that investors require a higher return to hold highly volatile securities. C. A portfolio of 75% in the risk free security and the remainder in the market has a beta of .75.
A. False - If the CAPM is true, they should offer the risk-free rate B. False - Only if the volatility increases the beta, which is not always the case C. False - It's 0.25
4. What is your response to a co-worker who says "If all securities are fairly priced, all must offer the same expected rates of return, otherwise the market is inefficient?"
Not true, because securities have different risks, thus different betas, thus different returns
Arbitrage is limited because the wealth of arbitrageurs is limited. Discuss this statement in the context of those who are managing their own money and those who are managing other people's money
Arbitrage is theoretically unlimited, but the amount of capital an arbitrageur has available is what limits profits. When they manage others money, they can realize more arbitrage profit, but they are still limited as they need to be more cautious when managing other's money, as their job is ultimately at the discretion of their clients.
What is the conjunctive fallacy and what are its implications for decision making?
Assuming that the probability of each of the parts is the same as the probability of the whole - not understanding compounding probabilities
Suppose an investor's utility function in wealth is U(w)=.1W-.00025W2 which the investor uses to maximize utility over the following choices:
E[U(X)] = .5(-13.256) + .5(-14.806) =-14.031E[U(Y)] = .5(-12.801) + .5(-15.921) =-14.361The investor will prefer X since it has a higher utility (smaller negative number).You do not use the Sharpe ratio because you know the utility function. The Sharpe ratio is only relevant when the CAPM is true and then you choose the market which will always have the highest Sharpe ratio) plus borrowing or lending.
Assume that the stock market consists of three stocks. Write down the formula for an equally weight index and a value weighted index of these three stocks: Stock A - Price: $20 - # of Shares: 1,000 Stock B - Price: $10 - # of Shares: 20,000 Stock C - Price: $40 - # of Shares: 5,000
Equally weighted index = (20 + 10 + 40)/3 Value weighted index = (20 x 1000) + (10 x 20000) + (40 x 5000)/440,000
Define hindsight bias, outcome bias, belief bias and confirmation bias. What are the sources of these biases?
Hindsight - we knew it all along Outcome - we let the outcome distort our view of the probabilities (very similar to hindsight) Belief - we have a narrative or frame that we want to interpret the world around, and as a result we only process information that supports our beliefs Confirmation - more active version of belief in that we actively seek information that supports our biases
IQ vs EQ
IQ - measure of your ability to solve problems and think logically EQ - measure of your ability to understand and manage emotions
How does stating a choice in the negative versus the positive affect which choice people make?
If you state a negative you are on the loss function, if you state a positive you are on the gain function
When eating out, Rory prefers spaghetti over a hamburger. Last night, she had a choice of spaghetti or macaroni and cheese and decided on the spaghetti again. The night before, Rory had a choice of spaghetti, pizza or a hamburger, and this time she had pizza. Then, today, she chose macaroni and cheese over a hamburger. Does her selection today indicate that Rory's choices are consistent with economic rationality? Why or why not?
No, it's not consistent with economic rationality bc her decision clearly fluctuates based on mood, circumstance, etc, rather than strictly going with her stated preference whenever possible.
Is your memory like a hard drive where you record an event and then recall it?
No, memory is constructive
What are the three supports on which market efficiency rests?
Investor rationality, uncorrelated errors and unlimited arbitrage.
Why does risk aversion require concave utility functions?
It is the only way you get a mean higher than the certainty equivalent - which results in a positive risk premium and is necessary for risk aversion.
Give two determinants of beta
Leverage and fixed assets - these make the stock more reactive to the market. Increasing leverage and fixed assets increases beta
What are the key variables that managers use to distinguish value stocks from growth stocks? What can we expect to earn by long value, short growth? What is the risk of this strategy?
Long value, short growth - you expect a positive return (this is HML). The risk is that value companies go out of business.
Define market efficiency
Market efficiency is the degree to which stock prices reflect all available and relevant information. It is not possible for an investor to outperform the market because all available information is already built into all stock prices. Broken into weak - no information Semi strong - public information Strong - all information
What is behavioral finance?
Marriage of finance and psychology that helps to explain irrational decision making in the market
Suppose you sell short 100 shares of IBM now selling at $120 per share. (See section 3.7 of BKM for short selling). What is your maximum possible loss? What happens to your maximum loss if you place a stop-buy order at $128?
Maximum possible loss is $12,000 (120 x 100). In principle, short sell losses are unlimited as stock price can rise infinitely. With the stop-buy order at $128, the maximum possible loss is $800 (100 x 8).
What do the mean, median and mode attempt to measure?
Mean - average (expected return) Median - middle number Mode - number that appears most often
Mood vs Emotion
Mood is how we feel over time and does necessarily depend on an event or trigger. Emotion is how we feel in each moment and is in response to something.
Is passive management identical with "buy and hold?"
No, but buy and hold is an example of a passive management strategy.
Warren Buffet has been a very successful investor. In 2008, Luisa Kroll reported that Buffet topped Forbes Magazine's list of the world's richest people with a fortune estimated to be worth $62 billion. Does this invalidate the EMH?
Not necessarily, as the EMH acknowledges that there will be exceptions. Additionally, the EMH argues that markets are efficient over the long term. So one could argue that Buffet is exploiting short-term inefficiencies.
What is the basic difference in how Kurtosis and skewness are computed? What do they attempt to measure?
The main difference is that X-bar is taking to the 3rd power for skewness, and to the 4th power for kurtosis. Negatively skewed distributions have a long left tail, which for investors can mean a greater chance of extremely negative outcomes. Positive skew would mean frequent small negative outcomes, and extremely bad scenarios are not as likely. Applied to investment returns, non symmetrical distributions are generally described as being either positively skewed (meaning frequent small losses and few extreme gains) or negatively skewed (meaning frequent small gains and a few extreme losses). Kurtosis refers to the degree of peak in a distribution. More peak than normal means that a distribution also has fatter tails and that there are less chances of extreme outcomes compared to a normal distribution.
What is the "Arbitrage Pricing Model?" What are the two assumptions and how is the arbitrage portfolio constructed?
The market model with two assumptions - that there is more than one factor and that arbitrage portfolio's earn close to a zero return in a well-functioning market
What is the St. Petersburg Paradox and why is it relevant for expected utility?
The paradox is why are people not willing to pay an infinite amount? Because they are willing to pay not the expected dollar amount, but the expected utility of the dollar amount
The Siegel data in the book "Stocks for the Long Run" shows that a well-diversified portfolio of stocks outperformed bonds over every thirty year period in the past 200 years. Further it shows that the standard deviation of stock returns is lower than bonds for twenty and thirty year periods. The data has not been seriously questioned and show that stocks are less risky than bonds. So why is there an equity risk premium?
The risk premium could be because of prospect theory - investors are extremely risk averse because stocks are more volatile over shorter periods such as 5 or 10 years. Or they do not believe the data. Most investors also likely have an outlook that is less than 30 years.
What is the difference between the sample mean (or sample median, sample mode) and the population mean (median or mode)?
The term population represents all possible measurements or outcomes that are of interest to us in a particular study. The term sample refers to a portion of the population that is representative of the population from which it was selected. In finance, we take a time series and assume it's from a distribution and predict that the series will continue in the future --analyze data based on this series as if it were representative of the entire population.
How is the market model different from the CAPM?
They both are one-factor models, but the market model has a regression assumption, whereas the CAPM has many more assumptions
What benefits do short-sellers provide for securities markets?
They offset optimism by uncovering negative information such as fraud. They also provide the ability to arbitrage.
How might an investor's portfolio have changed from 1995 to 2000 if the investor became overconfident?
They would be overweighted in internet stocks
How does the value philosophy differ from the growth philosophy?
Value is a representative bias (generally bc they are boring, not very popular, etc) so the market overlooks them and depresses the price. Signified by low P/E. Growth companies are exciting, explosive, etc and the popularity of the stock is based on the expected future performance of the company.
What does variance and standard deviation measure? What investment (really behavioral) assumption do they make?
Variance--Measure of dispersion of the distribution. Measure of risk. Standard Deviation--The square root of the variance - the only relevant measure of risk. The behavioral assumption is that investors care about the upside as much as the downside and weight them the same - when in reality investors don't typically care about positive variance, they care about negative variance.
How does distortion of the probabilities matter in prospect theory? (compare problems 3 and 4 in Behavioral Finance chapter 3 be sure you understand the notation.)
Very low probability events are overstated and very high probability events are understated
What is prospect theory?
We dislike losing more than we like winning - but the gain from winning and the hurt from losing decreases as we win/lose more
How does online trading contribute to the relationship between a portfolio's return and stock trading?
We trade too much because of how easy it is and we have the allusion of control
What is the joint hypothesis problem? Why is it important?
When testing for efficient markets, anomalous returns may reflect market inefficiencies, but it also may reflect an inaccurate asset pricing model or both. It's important bc it exemplifies how difficult it is to test for market efficiency, bc it's extremely difficult to gauge whether abnormalities stem from market inefficiency or inaccurate models.
Shares of small firms with thinly traded stocks tend to show positive CAPM alphas. Is this a violation of the efficient market hypothesis?
Yes
Can a one factor model be an "Arbitrage Pricing Model"? If so then what exactly is the difference between a one factor arbitrage pricing model and the CAPM?
Yes, and the difference depends on the factors. If the factor is the market, it is the CAPM. If the factor is inflation, employment, etc, it's not the CAPM
Suppose you find that stock prices before large dividend increases show significantly positive returns on average. Does this violate the Efficient Market Hypothesis?
Yes, it does violate the strong form efficient market hypothesis
When you use the historical mean and variance as measures of expected return and risk what assumptions are you making?
You make the assumption that the past can predict the future.
Suppose you are an active money manager and your clients ask you to choose a broad based index to be your benchmark. (You will probably respond broad based indices don't measure what I am doing, and your client responds "I don't care, I just want you to beat the market"). If the Russell 3000, SP500 and Wilshire 5000 have a correlation of .99, which index would you use and why? Will your answer be the same if you are a well-informed (about investments) client?
You would probably use the index with the lowest average return as a benchmark, making it easier for you to "beat the market." If you were a client, you would instead choose the one with the highest return as you want maximum performance out of your manager.
Differentiate the following terms/concepts: a. Primacy and recency effects b. Salience and availability c. Fast-and-frugal heuristics and bias-generating heuristics d. Autonomic and cognitive heuristics
a. Information learned at the beginning (primacy) and end (recency) are recalled better than information in the middle. b. Items or information that are especially remarkable (salient) received inflated estimates of probability. Information that is easily recalled and has many examples (availability) is also given inflated estimates of probability. c. Fast and frugal heuristics are simple rules of thumb that we use to interpret information and make decisions based off of it. Bias generating heuristics are when we incorrectly interpret a heuristic and draw a false conclusion. Fast and frugal heuristics can lead to bias-generating heuristics. d. Autonomic is system 1, cognitive is system 2
Differentiate the following terms/concepts: a. Miscalibration and excessive optimism b. Better-than-average effect and illusion of control c. Self-attribution bias and confirmation bias
a. Miscalibration is overestimating the precision of ones own forecast, whereas excessive optimism is overestimating our chance of positive experiences and underestimating our chance of negative experiences. They are both forms of overconfidence. b. BTAE is the tendency for people to perceive their abilities as superior to their peers. Illusion of control bias is believing we have more control over a situation than we actually do. c. Self-attribution bias is attributing success to ourselves and blaming failure on external factors. Confirmation bias is looking for or interpreting information in a way that supports our beliefs.
Differentiate the following terms/concepts: a. Momentum and reversal b. Value and growth stocks c. Fundamental risk and noise-trader risk d. Carve-out and stub value
a. Momentum trading is trying to enter a trend as it is picking up steam (buy high, sell higher). Reversal trading is when the price changes directions. Momentum traders try to exit their position before the reversal. b. Value is a boring company undervalued by the market. Growth is an exciting company that is expected to continue to outperform. c. Fundamental risk is "real risk" that can affect the price of a stock, such as interest rates, natural disasters, etc. Noise trader risk is the risk of uninformed investors that exhibit herd mentalities, and can impact the price of a stock. d. Carve out is a company spinning off a subsidiary as a standalone company. Stub value is the difference between the market value of the parent company after the carve out and the market value of the parent company's equity holdings in the carved out firm.
Differentiate the following terms/concepts: a. Lottery and insurance b. Segregation and integration c. Risk aversion and loss aversion d. Weighting function and event probability
a. Paying to possibly win, paying to not lose b. Together, separated c. Risk aversion is a general bias towards safety and away from the potential for loss, whereas loss aversion is that investors often want to avoid losing more than they want to win d. Weighting function is a mathematical tool to compute weighted average of elements in a set. Event probability is a tool to find out expected outcomes of an event.
Differentiate the following terms/concepts: a. Prospect and probability distribution b. Risk and uncertainty c. Utility function and expected utility d. Risk aversion, risk seeking and risk neutrality
a. Prospect describes all possible outcomes and how likely they are to happen. Probability distribution doesn't tell you what the outcomes mean, only how likely they are. b. Risk is when you have a good idea of what might happen and can estimate how likely each outcome is. Uncertainty is when you don't know what might happen and it's difficult to guess the chances of different outcomes. Risk is like rolling a die, uncertainty is like trying to guess the weather. c. Your utility function tells you how much you like things, and expected utility helps you make smart choices by considering what you like and the odds of things happening d. Don't like risk, like risk, neutral towards risk
Differentiate the following terms/concepts: a. Systematic and nonsystematic risk b. Beta and standard deviation c. Direct and indirect agency costs d. Weak, semi-strong, and strong form market efficiency
a. Systematic is big-picture risk that affects everyone, while nonsystematic risk is the individual, company-specific risk b. Beta measures how an investment behaves compared to the market, while standard deviation measures how much an investment's returns jump around by themselves c. Direct agency costs are the obvious expenses incurred to keep an eye on managers (auditing costs, etc), while indirect agency costs are the less obvious, sometimes unforeseen, negative outcomes that can arise from conflicts of interest between managers and shareholders d. Weak - no information priced in. Semi-strong - public information priced in. Strong - all information priced in.
What are the three key assumptions of an inefficient market? If the market is never efficient for a particular set of information can you still make money?
irrational, correlated, predictable