Kendall's Rank Correlation Coefficient
- To: mathgroup at smc.vnet.net
- Subject: [mg34685] Kendall's Rank Correlation Coefficient
- From: "Aaron E. Hirsh" <aehirsh at stanford.edu>
- Date: Fri, 31 May 2002 04:30:18 -0400 (EDT)
- Sender: owner-wri-mathgroup at wolfram.com
I've noticed that mathematica's calculation of Kendall's Rank Correlation Coefficient becomes unworkably inefficient as sample size grows. Spearman's Rank Correlation, by contrast, causes no problems. I've written a short mathematica program to calculate Kendall's coefficient, and it seems to be much faster than the built-in mathematica function. Does anyone know what the built-in function is doing that's taking so long? -- Aaron E. Hirsh Center for Computational Genetics and Biological Modeling Department of Biological Sciences Stanford University tel. (650) 723-4952 fax. (650) 725-0180