ABSTRACT

Ordinarily, a bivariate (or generally a multi-variable) design problem must be approached using the appropriate bivariate (multivariate) distribution. In the above example, the bivariate distribution for sitting knee height and sitting stature would be needed. Such a bivariate distribution for normally distributed variables would be expressed as the following equation in which if “X” stands for the sitting knee height then Y stands for the sitting head height, XY is the correlation betwen the two variables and “m” and “” stand for the means and standard deviations of the variates.