Keywords - Function groups - @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Library: stats
See also: relation relationchi2 relationcont relationcorrcont relationcorr

Quantlet: relationrank
Description: Computes the rank correlation of spearman for ordinal data.

Usage: rel = relationrank (x {, colname {, opt}})
Input:
x n x p variables
colname p x 1 text vector of names of x variables
opt q x 1 text vector of optional parameters
Output:
rel.r p x p matrix of rank correlation coefficients
rel.pval p x p unused (filled with NaNs)

Note:

Example:
; loads the library stats
library("stats")
; read swiss banknote data
x = read("bank2")
; compute the rank correlation coefficients
colname = string("X%0.f", 1:cols(x))
rel = relationrank(x, colname, "automatic")
rel.r

Result:
Contents of r;
[1,]        1   0.2256  0.10803 -0.18962 -0.050143  0.22878
[2,]   0.2256        1  0.75063  0.42721  0.36966 -0.48552
[3,]  0.10803  0.75063        1  0.52009  0.42863 -0.50498
[4,] -0.18962  0.42721  0.52009        1  0.16893 -0.60763
[5,] -0.050143  0.36966  0.42863  0.16893        1 -0.58344
[6,]  0.22878 -0.48552 -0.50498 -0.60763 -0.58344        1



Author: S. Klinke, 19970820
(C) MD*TECH Method and Data Technologies, 05.02.2006