Title:
Approximation of Cramer-von Mises test statistic
by weighted sum of chi-square variables.

Abstract:
   Making use of an elementary property of order statistics, it is shown
that Cramer-von Mises goodness of fit test statistics can be written as a
weighted sum of chi-squared variables for reasonably large sample.
It holds true even when parameters are unknown and estimated by
various estimation procedures under mild regularity conditions.