sparsity: Estimate sparsity of given matrix

In Columbia-PRIME/pcpr: Principal Component Pursuit for Environmental Epidemiology

Estimate sparsity of given matrix

Description

sparsity() estimates the percentage of entries in a given data matrix D whose values are "practically zero". If the absolute value of an entry is below a given threshold parameter thresh, then that value is determined to be "practically zero", increasing the estimated sparsity of D. Note that NA values are imputed as 0 before the sparsity calculation is made.

Usage

sparsity(D, thresh = 1e-04)

Arguments

D	The input data matrix.
thresh	(Optional) A numeric threshold ⁠>= 0⁠ used to determine if an entry in D is "practically zero". If the absolute value of an entry is below thresh, then it is judged to be "practically zero". By default, thresh = 1e-04.

Value

The sparsity of D, measured as the percentage of entries in D that are "practically zero".

See Also

matrix_rank()

Examples

sparsity(matrix(rep(c(1, 0), 8), 4, 4))
sparsity(matrix(0:8, 3, 3))
sparsity(matrix(0, 3, 3))

Columbia-PRIME/pcpr documentation built on April 14, 2025, 8:33 a.m.