pchc (version 1.2)

Partial correlation between two continuous variables: Partial correlation

Description

Partial correlation between two continuous variables when a correlation matrix is given.

Usage

pcor(R, indx, indy, indz, n)

Value

A numeric vector containing the partial correlation and logged p-value for the test of no partial correlation.

Arguments

R

A correlation or covariance matrix.

indx

The index of the first variable whose conditional correlation is to estimated.

indy

The index of the second variable whose conditional correlation is to estimated.

indz

The index of the conditioning variables.

n

The sample size of the data from which the correlation matrix was computed.

Author

Michail Tsagris.

R implementation and documentation: Michail Tsagris .

Details

Given a correlation or a covariance matrix the function will caclulate the partial correlation between variables indx and indy conditioning on variable(s) indz and will return the logarithm of the p-value.

See Also

cor2pcor, cortest, correls

Examples

y <- as.matrix( iris[, 1:2] )
z <- cbind(1, iris[, 3] )
er <- resid( .lm.fit(z, y) )
r <- cor(er)[1, 2]
z <- 0.5 * log( (1 + r) / (1 - r) ) * sqrt( 150 - 1 - 3 )
log(2) + pt( abs(z), 150 - 1 - 3, lower.tail = FALSE, log.p = TRUE )
r <- cor(iris[, 1:3])
pcor(r, 1,2, 3, 150)