Analyzing Stock Market Volatility Using Extreme-Day Measures

We develop a simple measure of volatility based on extreme-day returns and apply it to market returns during 1885 - 2002. Because returns are not normally distributed, the extreme-day measure, which is distribution free, might provide a better measure of stock market risk than the traditional standard deviation. The extreme-day measure more accurately explains investor behavior relative to standard deviation as shown by equity fund flows, and we find evidence that large negative changes appear to influence investor behavior more than large positive changes