Rank correlation matrix in R -

how produce rank correlation matrix in elegant way in r given data frame many columns? couldn't find built-in function, tried

> test=data.frame(x=c(1,2,3,4,5), y=c(5,4,3,2,1)) > cor(rank(test)) 

(only 2 columns simplicity, real data has 5 columns) gave

> error in cor(rank(test)) : supply both 'x' , 'y' or matrix-like 'x' 

i figured because rank takes single vector. tried

> cor(lapply(test,rank)) 

to rank applied each column in data frame, treating data frame list, gave error

> supply both 'x' , 'y' or matrix-like 'x' 

and ended getting working with

> cor(data.frame(lapply(test,rank)))    x  y x  1 -1 y -1  1 

however seems pretty verbose , ugly. i'm thinking there must better way -- if what?

you doing wrong -- use kendall method argument cor() instead:

r> testdf <- data.frame(x=c(1,2,3,4,5), y=c(5,4,3,2,1))   r> cor(testdf, method="kendall")     x  y  x  1 -1     y -1  1    r>  

from help(cor):

for cor(), if method "kendall" or "spearman", kendall's tau or spearman's rho statistic used estimate rank-based measure of association. these more robust , have been recommended if data not come bivariate normal distribution. cov(), non-pearson method unusual available sake of completeness. note "spearman" computes cor(r(x), r(y)) (or cov(.,.)) r(u) := rank(u, na.last="keep"). in case of missing values, ranks calculated depending on value of use, either based on complete observations, or based on pairwise completeness reranking each pair.