Skip to contents

C++ implementation of multivariate normal probability density function for multiple inputs

Source: R/RcppExports.R
mvnpdfC.Rd

Based on the implementation from Nino Hardt and Dicko Ahmadou https://gallery.rcpp.org/articles/dmvnorm_arma/ (accessed in August 2014)

Usage

mvnpdfC(x, mean, varcovM, Log = TRUE)

Arguments

x

data matrix

mean

mean vector

varcovM

variance covariance matrix

Log

logical flag for returning the log of the probability density function. Default is TRUE

Value

vector of densities

Author

Boris P. Hejblum

Examples

mvnpdf(x=matrix(1.96), mean=0, varcovM=diag(1), Log=FALSE)
#> [1] 0.05844094
mvnpdfC(x=matrix(1.96), mean=0, varcovM=diag(1), Log=FALSE)
#> [1] 0.05844094
mvnpdf(x=matrix(1.96), mean=0, varcovM=diag(1))
#> [1] -2.839739
mvnpdfC(x=matrix(1.96), mean=0, varcovM=diag(1))
#> [1] -2.839739

if(require(microbenchmark)){
library(microbenchmark)
microbenchmark(dnorm(1.96),
               mvnpdf(x=matrix(1.96), mean=0, varcovM=diag(1), Log=FALSE),
               mvnpdfC(x=matrix(1.96), mean=0, varcovM=diag(1), Log=FALSE),
               times=10000L)
}else{
cat("package 'microbenchmark' not available\n")
}
#> Unit: nanoseconds
#>                                                                 expr   min
#>                                                          dnorm(1.96)   791
#>   mvnpdf(x = matrix(1.96), mean = 0, varcovM = diag(1), Log = FALSE) 51596
#>  mvnpdfC(x = matrix(1.96), mean = 0, varcovM = diag(1), Log = FALSE)  4308
#>     lq      mean  median    uq     max neval
#>    931  1264.260  1302.0  1493   13836 10000
#>  54241 59183.584 55198.5 56746 2942525 10000
#>   4779  6313.462  5861.0  6372 2776827 10000