Friends I am new to R (and statistics) so am struggling a bit. Briefly... I am interested in getting the P-Value from cor(X) where X is a matrix. I have found cor.test. Verbosely... I have 4 vectors and can generate the corellation matrix...> cor(cbind(X1, X2, X3, X4))X1 X2 X3 X4 X1 1.00000000 -0.06190365 -0.156972795 0.182547517 X2 -0.06190365 1.00000000 0.264352860 0.146750844 X3 -0.15697279 0.26435286 1.000000000 -0.006380052 X4 0.18254752 0.14675084 -0.006380052 1.000000000 But I want the P-Values (gives me the significance I belive).> cor.test(X2, X3)Pearson's product-moment correlation data: X2 and X3 t = 3.3346, df = 148, p-value = 0.001080 alternative hypothesis: true correlation is not equal to 0 95 percent confidence interval: 0.1086963 0.4073565 sample estimates: cor 0.2643529 is very cool. Just what I want. But there is too much information for too little data. What I would like to do is some thing like... cor_with_p_test(cbind(X1, X2, X3, X4)) X1 X2 X3 X1 1 -0.06190365 -0.1569728... P 0.4517 0.05507 ... X2 -0.06190365 1 0.2643529 ... P 0.4517 0.001080 ... : : I think I could write a function for it if such a function does not exist if I could do...> CT23 <- cor.test(X2, X3) > CT23$P0.001080> CT23$V0.2643529 But I do not know how. cheers Worik
These functions are based on posts to either R-Help or S-News by
Gabor Grothendieck and Bill Venables.
# pairwise sample size
# Gabor G - 11/23/2004 R-help List
pn <- function(X){crossprod(!is.na(X))}
cor.prob <- function(X){
# Correlations Below Main Diagonal
# Significance Tests with Pairwise Deletion
# Above Main Diagonal
# Believe part of this came from Bill Venables
pair.SampSize <- pn(X)
above1 <- row(pair.SampSize) < col(pair.SampSize)
pair.df <- pair.SampSize[above1] - 2
R <- cor(X, use="pair")
above2 <- row(R) < col(R)
r2 <- R[above2]^2
Fstat <- (r2 * pair.df)/(1 - r2)
R[above2] <- 1 - pf(Fstat, 1, pair.df)
R
}
mydata <- matrix(rnorm(1000), ncol=10)
cor.prob(mydata)
Worik Turei stanton wrote:> Friends
>
> I am new to R (and statistics) so am struggling a bit.
>
> Briefly...
>
> I am interested in getting the P-Value from cor(X) where X is a matrix.
>
> I have found cor.test.
>
> Verbosely...
>
> I have 4 vectors and can generate the corellation matrix...
>
>
>
>>cor(cbind(X1, X2, X3, X4))
>
> X1 X2 X3 X4
> X1 1.00000000 -0.06190365 -0.156972795 0.182547517
> X2 -0.06190365 1.00000000 0.264352860 0.146750844
> X3 -0.15697279 0.26435286 1.000000000 -0.006380052
> X4 0.18254752 0.14675084 -0.006380052 1.000000000
>
> But I want the P-Values (gives me the significance I belive).
>
>
>>cor.test(X2, X3)
>
>
> Pearson's product-moment correlation
>
> data: X2 and X3
> t = 3.3346, df = 148, p-value = 0.001080
> alternative hypothesis: true correlation is not equal to 0
> 95 percent confidence interval:
> 0.1086963 0.4073565
> sample estimates:
> cor
> 0.2643529
>
> is very cool. Just what I want. But there is too much information for
> too little data.
>
> What I would like to do is some thing like...
>
> cor_with_p_test(cbind(X1, X2, X3, X4))
> X1 X2 X3
> X1 1 -0.06190365 -0.1569728...
> P 0.4517 0.05507 ...
>
> X2 -0.06190365 1 0.2643529 ...
> P 0.4517 0.001080 ...
> :
> :
>
> I think I could write a function for it if such a function does not exist
> if I could do...
>
>
>>CT23 <- cor.test(X2, X3)
>>CT23$P
>
> 0.001080
>
>>CT23$V
>
> 0.2643529
>
> But I do not know how.
>
> cheers
> Worik
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
>
--
Chuck Cleland, Ph.D.
NDRI, Inc.
71 West 23rd Street, 8th floor
New York, NY 10010
tel: (212) 845-4495 (Tu, Th)
tel: (732) 452-1424 (M, W, F)
fax: (917) 438-0894
Worik Turei stanton wrote:> Friends > > I am new to R (and statistics) so am struggling a bit. > > Briefly... > > I am interested in getting the P-Value from cor(X) where X is a matrix. > > I have found cor.test. > > Verbosely... > > I have 4 vectors and can generate the corellation matrix... > > > >>cor(cbind(X1, X2, X3, X4)) > > X1 X2 X3 X4 > X1 1.00000000 -0.06190365 -0.156972795 0.182547517 > X2 -0.06190365 1.00000000 0.264352860 0.146750844 > X3 -0.15697279 0.26435286 1.000000000 -0.006380052 > X4 0.18254752 0.14675084 -0.006380052 1.000000000 > > But I want the P-Values (gives me the significance I belive). > > >>cor.test(X2, X3) > > > Pearson's product-moment correlation > > data: X2 and X3 > t = 3.3346, df = 148, p-value = 0.001080 > alternative hypothesis: true correlation is not equal to 0 > 95 percent confidence interval: > 0.1086963 0.4073565 > sample estimates: > cor > 0.2643529 > > is very cool. Just what I want. But there is too much information for > too little data. > > What I would like to do is some thing like... > > cor_with_p_test(cbind(X1, X2, X3, X4)) > X1 X2 X3 > X1 1 -0.06190365 -0.1569728... > P 0.4517 0.05507 ... > > X2 -0.06190365 1 0.2643529 ... > P 0.4517 0.001080 ... > : > : > > I think I could write a function for it if such a function does not exist > if I could do... > > >>CT23 <- cor.test(X2, X3) >>CT23$P > > 0.001080 > >>CT23$V > > 0.2643529 > > But I do not know how. > > cheers > Woriklibrary(Hmisc) rcorr(cbind(. . . .)) -- Frank E Harrell Jr Professor and Chair School of Medicine Department of Biostatistics Vanderbilt University