centmat: Calculate a centralized version of the distance matrix.

In jofie/DistCov: Functions for distance correlations

Description

Calculate a centralized version of the distance matrix.

Usage

centmat(X, metr.X = "euclidean", type.X = "sample", bias.corr = TRUE,
  n, p, ...)

Arguments

X	a numeric vector or a numeric matrix.
metr.X	metric that should be used to compute the distance matrix.
bias.corr	logical; indicates if the corresponding estimator of the covariance matrix is biased or unbiased.
...	additional parameters that are used for other metrics (e.g., the bandwidth for Gaussian kernels)

Details

For metr.X the following metrices are built in: euclidean, gaussian and discrete. However, it is possible to use a function taking two numerical arguments as metr.X.

Value

The centralized distance matrix corresponding to X.

jofie/DistCov documentation built on May 23, 2019, 9:02 p.m.