increase_na_percentage {topolow} | R Documentation |
Increase Missing Values in a Matrix
Description
Strategically introduces NA values into a distance matrix while maintaining symmetry. New NA values are added preferentially farther from the diagonal to simulate real-world measurement patterns where distant pairs are more likely to be unmeasured.
Usage
increase_na_percentage(mat, target_na_percentage)
Arguments
mat |
Matrix to modify |
target_na_percentage |
Numeric between 0 and 1 specifying desired proportion of NAs |
Details
The function:
Calculates needed additional NAs to reach target percentage
Creates probability matrix favoring off-diagonal elements
Randomly selects positions weighted by distance from diagonal
Maintains matrix symmetry by mirroring NAs
Value
A matrix
with an increased number of NA
values, maintaining symmetry.
Examples
# Create sample distance matrix
dist_mat <- matrix(runif(100), 10, 10)
dist_mat[lower.tri(dist_mat)] <- t(dist_mat)[lower.tri(dist_mat)]
diag(dist_mat) <- 0
# Increase NAs to 70%
sparse_mat <- increase_na_percentage(dist_mat, 0.7)
[Package topolow version 1.0.0 Index]