I want to compute the Pvalue from the joint cumulative distribution of an ndimensional order statistic.
One efficient way is using the following recursive formula.
However, the facts are (or would be):
 I am too stupid to write a recursive function.
 I didn’t find the efficient formula at first.
 In other cases, the efficient formula have not been derived yet, or too complicated to derive.
In Statistics Monte Carlo simulation is a “quick” way to compute some complicated formulas. By saying “quick”, I mean I can see the results without knowing or deriving “ugly” Math formulas. It’s actually a very “slow” method in computing aspect.
Anyway, the R function is here.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 
