Double Centering of a Rectangular Matrix

This function performs double centering on a rectangular matrix. Double centering is a technique used to adjust a matrix so that both the row and column means are zero, even if the matrix is not square.

Usage

doubleCenterRect(x)

Arguments

x

A numeric rectangular matrix that you want to double center. The matrix can have different numbers of rows and columns.

Value

A numeric matrix of the same dimensions as x, with both rows and columns centered so that their means are zero.

Details

The function calculates the double-centered matrix by subtracting the row means, column means, and adjusting for the grand mean. The result is a matrix where both row and column means are zero.

Examples

if (FALSE) { # \dontrun{
# Load the example dataset
data("interest1981")

# Prepare the input matrix from columns 9 to 38
input <- as.matrix(interest1981[, 9:38])

# Set a cutoff for the minimum number of non-NA values per row
cutoff <- 5
input <- input[rowSums(!is.na(input)) >= cutoff, ]

# Transform the input data
input <- (100 - input) / 50

# Square the transformed input values
input2 <- input * input

# Replace NA values with the square of the mean of non-NA values
input2[is.na(input)] <- (mean(input, na.rm = TRUE))^2

# Perform double centering on the rectangular matrix
inputDC <- doubleCenterRect(input2)

# Print the double-centered matrix
print(inputDC)
} # }