Distribution functions for statistics derived from bivariate normal and bivariate two-parameter log-normal populations

Abstract

The distribution functions for hydrologic statistics that may be used to assess the significance of differences between sample means, standard deviations, coefficients of skewness, and coefficients of variations are obtained by Monte Carlo experiments. The distributions are expressed as functions of sample size, cross correlation, and skewness. In general, the distributions are more sensitive to cross correlation than to skewness. As sample size increases, however, the distributions tend to become more sensitive to skewness. (Woodard-USGS)