mosthighlycorrelated function error

[keithwlarson]

Running the following function from your book on multvariate analysis results in the below error (at the bottom).

mosthighlycorrelated <- function(mydataframe,numtoreport)
{

find the correlations

cormatrix <- cor(mydataframe)

set the correlations on the diagonal or lower triangle to zero,

so they will not be reported as the highest ones:

diag(cormatrix) <- 0
cormatrix[lower.tri(cormatrix)] <- 0

find the dimensions of the matrix, and the row names:

numrows <- nrow(cormatrix)
therownames <- rownames(cormatrix)

find the highest correlations

sorted <- sort(abs(cormatrix),decreasing=TRUE)
for (i in 1:numtoreport)
{
corri <- sorted[i]

find the pair of variables with this correlation

for (j in 1:(numrows-1))
{
for (k in (j+1):numrows)
{
corrjk <- cormatrix[j,k]
if (corri == abs(corrjk))
{
rowname <- therownames[j]
colname <- therownames[k]
print(paste("i=",i,"variables",rowname,"and",colname,"correlation=",corrjk))
}
}
}
}
}

Error reported from r 2.14:
Error in if (corri == abs(corrjk)) { : missing value where TRUE/FALSE needed

[avrilcoghlan]

Dear Keith,

Thanks for letting me know this, I will fix it as soon as I can find time.

Kind Regards,
Avril

-----Original Message-----
From: Keith Larson [mailto:[email protected]]
Sent: Sun 12/11/2011 3:09 PM
To: Coghlan, Avril
Subject: [LittleBookofRMultivariateAnalysis] mosthighlycorrelated function error (#1)

Running the following function from your book on multvariate analysis results in the below error (at the bottom).

mosthighlycorrelated <- function(mydataframe,numtoreport)
{

find the correlations

cormatrix <- cor(mydataframe)

set the correlations on the diagonal or lower triangle to zero,

so they will not be reported as the highest ones:

diag(cormatrix) <- 0
cormatrix[lower.tri(cormatrix)] <- 0

find the dimensions of the matrix, and the row names:

numrows <- nrow(cormatrix)
therownames <- rownames(cormatrix)

find the highest correlations

sorted <- sort(abs(cormatrix),decreasing=TRUE)
for (i in 1:numtoreport)
{
corri <- sorted[i]

find the pair of variables with this correlation

for (j in 1:(numrows-1))
{
for (k in (j+1):numrows)
{
corrjk <- cormatrix[j,k]
if (corri == abs(corrjk))
{
rowname <- therownames[j]
colname <- therownames[k]
print(paste("i=",i,"variables",rowname,"and",colname,"correlation=",corrjk))
}
}
}
}
}

Error reported from r 2.14:
Error in if (corri == abs(corrjk)) { : missing value where TRUE/FALSE needed

---

Reply to this email directly or view it on GitHub:
#1

[avrilcoghlan]

Dear Keith,

I've finally got around to looking at this (sorry about the delay). I'm using 2.15.1 and don't get this error using this function on the wine[2:14] data-frame in the example on the webpage. Is this the input you are using?
If not, please could you send me the input, so I can try to replicate the error?

Regards,
Avril

[zkanfra]

So how was this fixed. I have the same error message