Programming the statistical procedures from SAS

Calculation of Kolmogorov-Smirnov p-value in NPAR1WAY

Reply
Occasional Contributor
Posts: 14

Calculation of Kolmogorov-Smirnov p-value in NPAR1WAY

What exact algorithm is used to calculate the asymptotic K-S p-value given the d-statistic and n (or n1 and n2 for two-sided tests) in NPAR1WAY?  The documentation I found shows the theoretical equations, which involve an infinite sum, but my understanding is that in practice approximations are always used (at minimum the infinite sum must be truncated).

Also, is the function that does the calculation available to be called in by users in a data step (as is the case in R and other stat packages)? 

Respected Advisor
Posts: 4,606

Re: Calculation of Kolmogorov-Smirnov p-value in NPAR1WAY

Did you come across some incorrect p-values with proc NPAR1WAY?

PG
Occasional Contributor
Posts: 14

Re: Calculation of Kolmogorov-Smirnov p-value in NPAR1WAY

Within the scope of what NPAR1WAY addresses, no problems, but NPAR1WAY does NOT work on weighted data. 

I had to fool it by using FREQ as the weight and recalculating the p-value from _d_, n1 and n2 using a function I wrote myself.  To my surprise, for unweighted data my calculation and the one by NPAR1WAY agree to 8 significant digits!  I suspect I implemented the very same algorithm for the calculation that NPAR1WAY uses, which is why I'm wondering...

Ask a Question
Discussion stats
  • 2 replies
  • 282 views
  • 0 likes
  • 2 in conversation