dmvlogis: Density of the Multivariate Logistic Distribution
In LocaTT: Geographically-Conscious Taxonomic Assignment for Metabarcoding

dmvlogis

R Documentation

Density of the Multivariate Logistic Distribution

Description

Density function for the multivariate logistic distribution.

Usage

dmvlogis(x, location, scale, R, log = FALSE)

Arguments

`x`	Numeric vector or matrix. Values of logistically-distributed marginals. If matrix, then a vector of probability densities is returned with an element for each record of the matrix. Matrix records represent observations, and matrix fields represent dimensions.
`location`	Numeric vector. Location parameters of the logistic distribution. If `x` is a matrix, then `location` is recycled for each record of matrix `x`.
`scale`	Numeric vector. Scale parameters of the logistic distribution. If `x` is a matrix, then `scale` is recycled for each record of matrix `x`.
`R`	Numeric correlation matrix. If `x` is a matrix, then `R` is recycled for each record of matrix `x`.
`log`	Logical scalar. If `TRUE`, then probabilities are given as `log(density)`.

Details

Computes the probability density of the multivariate logistic distribution. The multivariate logistic distribution is constructed using a Gaussian copula with logistic marginals. The probability density is the product of the densities of the logistic marginals, which is further multiplied by the density of a Gaussian copula of the transformed standard uniform margins (i.e., probability integral transformation of the logistic marginals with stats::plogis).

Value

Numeric vector of probability densities.

References

Decani JS, and Stine RA. 1986. A note on deriving the information matrix for a logistic distribution. The American Statistician, 40(3): 220-222. DOI: 10.2307/2684541

Song P. 2000. Multivariate dispersion models generated from Gaussian copula. Scandinavian Journal of Statistics, 27(2): 305-320. DOI: 10.1111/1467-9469.00191

Examples

# Define logistic margins.
x<-c(0.055,-1.625,0.329,-5.765)

# Define location parameters.
location<-c(0.477,-0.998,-0.776,0.064)

# Define scale parameters.
scale<-c(0.574,1.314,0.460,1.393)

# Define correlation matrix.
R<-matrix(data=c(1.000,-0.80,0.64,-0.512,
                 -0.800,1.00,-0.80,0.640,
                 0.640,-0.80,1.00,-0.800,
                 -0.512,0.64,-0.80,1.000),
           ncol=4,byrow=TRUE)

# Compute log probability density.
dmvlogis(x=x,location=location,
         scale=scale,R=R,
         log=TRUE)

LocaTT documentation built on June 14, 2026, 1:06 a.m.