Applied Corporate Finance THE BASICS OF RISK参考.ppt

This is a summary of the CAPM, before we get into the details. Note that the variance that the CAPM is built around is the variance of actual returns around an expected return. If you were an investor with a 1-year time horizon, and you bought a 1-year T.Bill, your actual returns (at least in nominal terms) will be equal to your expected return. It is riskfree. If you were the same investor, and you bought a stock (say Intel), your actual returns will almost certainly not be equal to your expected returns. In practice, we often look at historical (past) returns to estimate variances. Implicitly, we are assuming that this variance is a good proxy for expected future variance. Disney’s stock price has been volatile, yielding a standard deviation of 19.36% (on an annualized basis) between 12004 and 2008. If you were an investor looking at Disney in 2009, what concerns (if any) would you have in using this as your measure of the forward looking risk in Disney stock? Disney as a company changed over this period. The standard deviation from the past may not be a good indicator or future risk. If historical standard deviations are your only way of estimating risk, it makes it impossible to measure risk in non-traded assets. While some people may be indifferent, most pick investment A. The possibility of a high payoff, even though it is captured in the expected value, seems to tilt investors. In statistical terms, this can be viewed as evidence that investors prefer positive skewness (high positive payoffs) and value it. It is a direct contradiction to the mean-variance framework that underlies so much of conventional risk theory. Once you add the possibility that the big positive jumps are matched by the possibility of big negative jumps, the game changes again. This propensity, called kurtosis, is not desirable to most investors. In the real world, investments reveal far too much skewness and kurtosis than would be expected in the standard normal distribution. In fact,