2.8 Alpha, Beta, and Hypothesis Testing Flashcards
Ex-Ante vs Ex-Post Alpha
Ex-Ante: EXPECTED consistent superior return generated through skill
Ex-Post: BACKWARD-LOOKING superior performance based on realised return. skill + luck
Simple Linear Regression
Y = a +bX + e
Ordinary Least Squares OLS Regression
Best fit of data points that minimises sum of squared error terms
Explain issues of OLS:
1. Outliers
2. Autocorrelation
3. Heteroskedasticity
- Outlier: Data sets with fat-tailed (Leptokurtic) distributions = higher error
- Autocorrelation: illiquid assets get smoothed or appraised praises
- Heteroskedasticity: variance varies over time (<,> shapes)
OLS Goodness of fit = r-squared
r-squared close to one means that the model is a good fit (i.e., variable X explains most of the variance in variable Y). Thus, all else equal, larger values of r-squared are preferred.
e.g 0.8 means that 80% of the returns/variation can be explained by systematic risk, therefore 20% alpha
T-test Statistical significance
t test = estimated value / standard error
at 5% level, t-statistic must exceed critical value of 1.96 for it to be statistically significant
In the context of hypothesis testing: Test Statistic =
(Estimated Value - Hypothesized Value) / Standard Error of Statistic
p value
Result generated by the statistical test that indicates the probability of obtaining a test statistic by chance that is equal to or more extreme than the one that was actually observed (under the condition that the null hypothesis is true).
The p-value that the test generated is then compared to the level of significance that the researcher chose.
Type I vs Type II errors in hypothesis testing
Type I: REJECT null hypothesis when it is true
Type II: FAIL TO REJECT when it is false
List the three major types of model misspecification in the context of estimating systematic risk.
Omitted (or misidentified) systematic return factors
Misestimated betas
Nonlinear risk-return relationships
Does ex ante alpha lead to ex post alpha?
Not necessarily. While ex ante alpha may be viewed as expected idiosyncratic return, ex post alpha is realized idiosyncratic return. Simply put, ex post alpha is the extent to which an asset outperformed or underperformed its benchmark in a specified time period. Ex post alpha can be the result of luck and/or skill. To the extent that an investor suffers bad luck, ex ante will not guarantee ex post alpha.
Does an analyst select a p-value or a significance level in preparation for a test?
Significance level. P Value is the output.
What are the two steps to an analysis of ex ante alpha using historical data?
An asset pricing model or benchmark must be used to divide the historical returns into the portions attributable to systematic risks (and the risk-free rate) and those attributable to idiosyncratic effects.
The remaining returns, meaning the idiosyncratic returns (i.e., ex post alpha), should be statistically analyzed to estimate the extent, if any, to which the superior returns may be attributable to skill rather than luck.
What is the goal of an empirical investigation of abnormal return persistence?
To identify ex ante alpha
Provide two common interpretations of the investment term alpha.
- Alpha refers to any excess or deficient investment return after the return has been adjusted for the time value of money (the risk-free rate) and for the effects of bearing systematic risk (beta).
- Alpha can also refer to the extent to which the skill, information, and knowledge of an investment manager generates superior risk-adjusted returns (or inferior risk-adjusted return in the case of negative alpha).
- Note that the first interpretation can include high returns from luck.