Freedom251 › Statistical Analysis Software
logKDE
What logKDE is
logKDE is a free R package for kernel density estimation on positive-domain data. Its estimators produce values only for positive inputs. The main logdensity function computes estimates directly, while logdensity_fft uses a fast Fourier transformation to speed computation. Available kernels are Gaussian, Epanechnikov, Laplace, logistic, triangular, and uniform. Bandwidth choices include methods applied to log-transformed data, positive-domain unbiased cross-validation, and a Silverman-type rule of thumb. Estimation outputs work with standard R plot commands. The package is available from CRAN, with instructions for installing the current GitHub build. CRAN lists Rcpp and pracma as imports and Rcpp as a linking dependency. Windows binaries and macOS binaries for arm64 and x86_64 are listed. logKDE is licensed under GPL-3. Authors are Hien D. Nguyen, Andrew T. Jones, and Geoffrey J. McLachlan; Andrew Thomas Jones is maintainer. The listed CRAN version is 0.3.3.
Who it suits
logKDE suits R users estimating densities for data restricted to positive values. It is useful when selectable kernels, bandwidth methods, plotting, and CRAN or GitHub installation are needed.
What it does well
- Designed for positive-domain data
- Offers direct and FFT estimation
- Six kernel choices listed
- Multiple bandwidth methods
- Available from CRAN
What to know first
- No statistical tests
- No regression models
- No Bayesian analysis
- Estimates only positive values
Plans
| Plan | Billed and limits | Price |
|---|---|---|
| logKDE | R package · GPL-3 | Free |
As andrewthomasjones.com.au lists them, read 11 Oct 2026. The maker's own page has the final word.
Verdict
logKDE focuses on positive-valued kernel density estimation in R, with direct and FFT-based functions plus several kernels and bandwidth options. It does not provide statistical tests, regression models, or Bayesian analysis.
Facts from logKDE's own pages
- Authors and maintainer
- The listed authors are Hien D. Nguyen, Andrew T. Jones, and Geoffrey J. McLachlan; Andrew Thomas Jones is listed as maintainer. · andrewthomasjones.com.au, 11 Oct 2026
- Bandwidth methods
- Bandwidth options include methods applied to log-transformed data, positive-domain unbiased cross-validation, and a Silverman-type rule of thumb. · andrewthomasjones.com.au, 11 Oct 2026
- Bandwidth selection
- Available bandwidth methods include methods used by R's density function on log-transformed data, positive-domain unbiased cross-validation, and a Silverman-type rule of thumb. · andrewthomasjones.com.au, 11 Oct 2026
- Bayesian analysis
- No · andrewthomasjones.com.au, 24 Sept 2026
- Dependencies
- CRAN lists Rcpp and pracma as imports, and Rcpp as a linking dependency. · cloud.r-project.org, 11 Oct 2026
- Deployment
- desktop · andrewthomasjones.com.au, 24 Sept 2026
- Estimation methods
- Its main functions are logdensity, which computes estimates directly, and logdensity_fft, which uses fast Fourier transformation to speed up computation. · andrewthomasjones.com.au, 11 Oct 2026
- Free plan
- Yes · andrewthomasjones.com.au, 24 Sept 2026
- Installation
- The package is available from CRAN, and the site also gives instructions for installing the current build from GitHub. · andrewthomasjones.com.au, 11 Oct 2026
- Intended data domain
- The estimator is designed for positive-domain data and produces estimates only over positive values. · andrewthomasjones.com.au, 11 Oct 2026
- Kernels
- Available kernels include Gaussian, Epanechnikov, Laplace, logistic, triangular, and uniform. · andrewthomasjones.com.au, 11 Oct 2026
- License
- The package is licensed under GPL-3. · cloud.r-project.org, 11 Oct 2026
- Main functions
- Its main functions are logdensity and logdensity_fft, with the latter using fast Fourier transformation to speed computation. · andrewthomasjones.com.au, 11 Oct 2026
- Package version
- CRAN lists version 0.3.3, published 2025-11-29. · cloud.r-project.org, 11 Oct 2026
- Platform binaries
- CRAN lists Windows binaries and macOS binaries for arm64 and x86_64. · cloud.r-project.org, 11 Oct 2026
- Platforms
- CRAN lists Windows binaries and macOS binaries for arm64 and x86_64. · cloud.r-project.org, 11 Oct 2026
- Plotting
- The estimation functions can be used with plot and produce output objects that support the usual plot commands. · andrewthomasjones.com.au, 11 Oct 2026
- Purpose
- logKDE provides R functions for kernel density estimation on positive data using log-kernel density functions. · andrewthomasjones.com.au, 11 Oct 2026
- Regression models
- No · andrewthomasjones.com.au, 24 Sept 2026
- Scripting support
- Yes · andrewthomasjones.com.au, 24 Sept 2026
- Statistical tests
- No · andrewthomasjones.com.au, 24 Sept 2026
- Support
- The maker's site directs bug reports and support requests to the GitHub Issues page. · andrewthomasjones.com.au, 11 Oct 2026
- Testing
- The maker's site reports 74 passing tests, with no failures, warnings, or skips, for the GitHub build tested on 29 November 2025. · andrewthomasjones.com.au, 11 Oct 2026
Where logKDE stands on Freedom251
- Statistical Analysis Software#56 of 106