Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: comparing tables of ranks


From   David Airey <david.airey@vanderbilt.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: comparing tables of ranks
Date   Sun, 14 Aug 2005 17:27:34 -0500

.

I'm trying to get,

1) a p value for a set of rankings occurring by chance, and

2) a p value for the correspondence between two sets of rankings.

The first data set has rankings of 3 molecular measures (relative abundance of protein isoforms) from 3 inbred (isogenic) mouse strains. In the table below, the 3 strains of mice are c57, dba, and balb.

strain ranks
--------------------
isoform c57 dba balb
------------------------------
1 1 3 2
2 1 3 2
3 1 2 3


The second data set has ranking of strains for 5 measures of behavior.

strain ranks
--------------------
behavior c57 dba balb
------------------------------
1 1 3 2
2 1 3 2
3 1 2 3
4 1 3 2
5 1 2 3

For example, regarding the p value for (1) above, I'd think that I could directly say that the probability of observing c57 being ranked "1" across 5 behavior measures would be, because there are 6 permutations for 3 objects,

(1,2,3 or 1,3,2)^5 = (1/6 + 1/6)^5 = 0.004

and for the top table,

(1,2,3 or 1,3,2)^3 = (1/6 + 1/6)^3 = 0.04

Is their a simple Stata command to get a p value for the correspondence between isoform and behavior for the strain ranks or should I develop a randomization test?

The ranks are based on the means of 10 different animals measured per strain, and different animals are in the two data sets.

Cheers,

Dave

*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/




© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index