This PR fixes issue #156 where the Chi-squared test statistic needed to be scaled by the number of degrees of freedom in order to yield asymptotically valid p-values from the U(.) function. This ensures the p-values from the Chi-square test will be asymptotically equivalent to those from the F-test.
If X follows an F distribution with q and n degrees of freedom, then Y = \lim_{n \to \infty} {q * X} follows a Chi-squared distribution with q degrees of freedom.
This PR fixes issue #156 where the Chi-squared test statistic needed to be scaled by the number of degrees of freedom in order to yield asymptotically valid p-values from the U(.) function. This ensures the p-values from the Chi-square test will be asymptotically equivalent to those from the F-test.
If X follows an F distribution with q and n degrees of freedom, then Y = \lim_{n \to \infty} {q * X} follows a Chi-squared distribution with q degrees of freedom.