GGMridge (version 1.4)

R.separate.ridge: Estimation of Partial Correlation Matrix Using p Separate Ridge Regressions.

Description

The partial correlation matrix is estimated by p separate ridge regressions with the parameters selected by cross validation.

Usage

R.separate.ridge(x, fold, lambda, verbose = FALSE)

Value

A list containing

R

The partial correlation matrix.

lambda.sel

The selected tuning parameters for p ridge regressions.

Arguments

x

n x p data matrix; n is the # of samples and p is the # of variables.

fold

Ridge parameters are selected by fold-cross validations separately for each regression.

lambda

The candidate ridge parameters for all p ridge regressions.

verbose

TRUE/FALSE; if TRUE, print the procedure.

Author

Min Jin Ha

References

Ha, M. J. and Sun, W. (2014). Partial correlation matrix estimation using ridge penalty followed by thresholding and re-estimation. Biometrics, 70, 762--770.

Examples

   p <- 100 # number of variables
   n <- 50 # sample size
   
   ###############################
   # Simulate data
   ###############################
   simulation <- simulateData(G = p, etaA = 0.02, n = n, r = 1)
   data <- simulation$data[[1L]]
   stddata <- scale(x = data, center = TRUE, scale = FALSE)
   
   ###############################
   # estimate ridge parameter
   ###############################
   w.upper <- which(upper.tri(diag(p)))
   
   lambda.array <- seq(from = 0.1, to = 20, by=0.1) * (n-1.0)
   partial.sep <-  R.separate.ridge(x = stddata,
                                    lambda = lambda.array,
                                    fold = 5L,
                                    verbose = TRUE)$R[w.upper]