R/FanPC_CoFM.R

Defines functions FanPC.CoFM

Documented in FanPC.CoFM

#' @name FanPC.CoFM
#' @title FanPC estimation for the Copula Factor Model
#' @description Perform FanPC (eigen-structure thresholding) on CoFM data to estimate loadings (AF), uniquenesses (DF) and their errors.
#' @param data  n×p data matrix (generated by \code{CoFM}).
#' @param m     Number of latent factors.
#' @param A     True p×m loading matrix (for error evaluation).
#' @param D     True p×p uniqueness matrix (for error evaluation).
#' @param p     Number of variables.
#' @return List with components
#'   \item{AF}{Estimated loadings.}
#'   \item{DF}{Estimated uniquenesses.}
#'   \item{MSESigmaA, MSESigmaD, LSigmaA, LSigmaD}{Error metrics (same definition as FanPC.SFM).}
#' @examples
#' \donttest{
#'   library(copula); library(MASS)
#'   X <- CoFM(n=1000, p=10, m=5, theta=2, distribution_type="Clayton")
#'   res <- FanPC.CoFM(X$data, m=5, X$A, diag(diag(X$D)), p=10)
#'   print(res)
#' }
#' @export
#' @importFrom matrixcalc frobenius.norm
#' @importFrom stats cor
FanPC.CoFM <- function(data, m, A, D, p) {
  if (!is.matrix(data)) data <- as.matrix(data)
  X  <- scale(data)
  R  <- cor(X)
  eig <- eigen(R)
  idx <- order(eig$values, decreasing = TRUE)[1:m]
  AF  <- eig$vectors[, idx] %*% diag(sqrt(eig$values[idx]))
  h2  <- rowSums(AF^2)
  DF  <- diag(R - diag(h2))

  MSESigmaA <- frobenius.norm(AF - A)^2 / (p^2)
  MSESigmaD <- frobenius.norm(DF - D)^2 / (p^2)
  LSigmaA   <- frobenius.norm(AF - A)^2 / frobenius.norm(A)^2
  LSigmaD   <- frobenius.norm(DF - D)^2 / frobenius.norm(D)^2
  list(AF = AF, DF = DF,
       MSESigmaA = MSESigmaA, MSESigmaD = MSESigmaD,
       LSigmaA   = LSigmaA,   LSigmaD   = LSigmaD)
}

Try the CoFM package in your browser

Any scripts or data that you put into this service are public.

CoFM documentation built on Nov. 18, 2026, 5:08 p.m.