How useful was Monte Carlo-type analysis in preparing for the recent downturn in the economy and stock market? Is there an alternative approach that investors should consider in an effort to address the uncertainty of future returns? EFF: Monte Carlo analysis is overkill here. All one really needs is good historical perspective on the volatility of volatility. Our white paper, "How Unusual Was the Stock Market of 2008?", is a good start. KRF: Monte Carlo analysis is often worse than overkill because it gives many users a false sense of precision. If used right, it can provide some perspective about the payoff on a long-term investment. Investors who do Monte Carlo simulations, however, often assume returns are drawn from a normal distribution with a constant volatility. In fact, a normal distribution produces far fewer extreme returns than we see in the market. Moreover, it is easy to forget that the inputs for the analysis - the estimated expected returns, variances, and covariances - are almost certainly grossly imprecise.
↧