Package: ksm 1.1

ksm: Kernel Density Estimation for Random Symmetric Positive Definite Matrices

Kernel smoothing for Wishart random matrices described in Daayeb, Khardani and Ouimet (2025) <doi:10.48550/arXiv.2506.08816>, Gaussian and log-Gaussian models using least square or likelihood cross validation criteria for optimal bandwidth selection.

Authors:Leo Belzile [aut, cre], Frederic Ouimet [aut]

ksm_1.1.tar.gz
ksm_1.1.zip(r-4.7)ksm_1.1.zip(r-4.6)ksm_1.1.zip(r-4.5)
ksm_1.1.tgz(r-4.6-x86_64)ksm_1.1.tgz(r-4.6-arm64)ksm_1.1.tgz(r-4.5-x86_64)ksm_1.1.tgz(r-4.5-arm64)
ksm_1.1.tar.gz(r-4.7-arm64)ksm_1.1.tar.gz(r-4.7-x86_64)ksm_1.1.tar.gz(r-4.6-arm64)ksm_1.1.tar.gz(r-4.6-x86_64)
ksm_1.1.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
ksm/json (API)

# Install 'ksm' in R:

install.packages('ksm', repos = c('https://lbelzile.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/lbelzile/ksm/issues

Uses libs:

openblas– Optimized BLAS
c++– GNU Standard C++ Library v3
openmp– GCC OpenMP (GOMP) support library

Datasets:

realvar - Realized variance of Amazon and SPY

On CRAN:

openblas cpp openmp

4.54 score 130 downloads 37 exports 2 dependencies

Last updated from:4ae6837fe1. Checks:13 OK. Indexed: yes.

Target	Result	Time
linux-devel-arm64	OK	129
linux-devel-x86_64	OK	127
source / vignettes	OK	193
linux-release-arm64	OK	121
linux-release-x86_64	OK	116
macos-release-arm64	OK	86
macos-release-x86_64	OK	261
macos-oldrel-arm64	OK	91
macos-oldrel-x86_64	OK	216
windows-devel	OK	118
windows-release	OK	112
windows-oldrel	OK	107
wasm-release	OK	113

Exports:bandwidth_optim dinvWishart dmbeta2 dsmlnorm dsmnorm dWishart integrate_spd kdens_smlnorm kdens_smnorm kdens_symmat kdens_Wishart lcv_kdens_symmat lcv_kern_smlnorm lcv_kern_smnorm lcv_kern_Wishart lscv_kdens_symmat lscv_kern_smlnorm lscv_kern_smnorm lscv_kern_Wishart mgamma Riccati rinvWishart rmbeta2 rmnorm rotation_scaling rotation2d rotation3d rVAR rWAR rWishart simu_fdens simu_ise_montecarlo simu_kldiv simu_rdens sumlog sumsignedlog symmetrize

Dependencies:Rcpp RcppArmadillo

Overview of the ksm package

Evaluation of the kernel and bandwidth optimization | Integrating over positive definite matrices

Rendered fromvignette_ksm.Rmdusingknitr::rmarkdown

Last update: 2026-06-06
Started: 2026-06-06

Citation

Development and contributors

Readme and manuals

Help Manual

Help page	Topics
Bandwidth optimization for symmetric matrix kernels	bandwidth_optim
Density of inverse Wishart random matrix	dinvWishart
Matrix beta type II density function	dmbeta2
Symmetric matrix-variate lognormal density	dsmlnorm
Symmetric matrix-variate normal density	dsmnorm
Density of Wishart random matrix	dWishart
Integration with respect to symmetric positive definite matrices	integrate_spd
Symmetric matrix log-normal kernel density	kdens_smlnorm
Symmetric matrix normal kernel density	kdens_smnorm
Kernel density estimators for symmetric matrices	kdens_symmat
Wishart kernel density	kdens_Wishart
Likelihood cross-validation for symmetric positive definite matrix kernels	lcv_kdens_symmat
Likelihood cross validation criterion for symmetric matrix lognormal kernel	lcv_kern_smlnorm
Likelihood cross validation criterion for symmetric matrix normal kernel	lcv_kern_smnorm
Likelihood cross validation criterion for Wishart kernel	lcv_kern_Wishart
Least square cross-validation for symmetric positive definite matrix kernels	lscv_kdens_symmat
Least square cross validation criterion for log symmetric matrix normal kernel	lscv_kern_smlnorm
Least square cross validation criterion for matrix normal kernel	lscv_kern_smnorm
Least square cross validation criterion for Wishart kernel	lscv_kern_Wishart
Multivariate gamma function	mgamma
Realized variance of Amazon and SPY	realvar
Solver for Riccati equation	Riccati
Random matrix generation from the inverse Wishart distribution	rinvWishart
Random matrix generation from matrix beta type II distribution	rmbeta2
Random vector generation from the multivariate normal distribution	rmnorm
Random matrix generation from first-order autoregressive Wishart process	rWAR
Random matrix generation from Wishart distribution	rWishart
Symmetrize matrix	symmetrize