COKURTOSIS

A statistical measure that calculates the degree of peak of a variable's probability distribution in relation to another variable's peakedness. All other things equal, a higher cokurtosis means that the first variable has a flatter probability distribution.

In finance, cokurtosis can be used as a supplement to the covariance calculation of risk estimation. Usually cokurtosis is calculated using a security's historic price data as the first variable, and the market's historic price data as the second. This provides an estimation of the security's risk in relation to the market.

For a risk-adverse investor, a lower cokurtosis is preferred, as the security's returns would not be much different from the market's returns (i.e. low beta).