Handles univariate non-parametric density estimation with parametric starts and asymmetric kernels in a simple and flexible way. Kernel density estimation with parametric starts involves fitting a parametric density to the data before making a correction with kernel density estimation, see Hjort & Glad (1995) <doi:10.1214/aos/1176324627>. Asymmetric kernels make kernel density estimation more efficient on bounded intervals such as (0, 1) and the positive half-line. Supported asymmetric kernels are the gamma kernel of Chen (2000) <doi:10.1023/A:1004165218295>, the beta kernel of Chen (1999) <doi:10.1016/S0167-9473(99)00010-9>, and the copula kernel of Jones & Henderson (2007) <doi:10.1093/biomet/asm068>. User-supplied kernels, parametric starts, and bandwidths are supported.
| Version: | 1.0.0 |
| Imports: | assertthat, EQL, knitr, rmarkdown |
| Suggests: | extraDistr, SkewHyperbolic, testthat, covr |
| Published: | 2018-02-27 |
| Author: | Jonas Moss, Martin Tveten |
| Maintainer: | Jonas Moss <jonas.gjertsen at gmail.com> |
| License: | MIT + file LICENSE |
| NeedsCompilation: | no |
| Materials: | README |
| CRAN checks: | kdensity results |
| Reference manual: | kdensity.pdf |
| Vignettes: |
Tutorial for 'kdensity' |
| Package source: | kdensity_1.0.0.tar.gz |
| Windows binaries: | r-devel: kdensity_1.0.0.zip, r-release: kdensity_1.0.0.zip, r-oldrel: kdensity_1.0.0.zip |
| OS X binaries: | r-release: kdensity_1.0.0.tgz, r-oldrel: kdensity_1.0.0.tgz |
Please use the canonical form https://CRAN.R-project.org/package=kdensity to link to this page.