Assess the significance of identified clusters and estimates the true number of clusters by comparing the explained variation due to the clustering from the original data to that produced by clustering a unimodal reference distribution which preserves the covariance structure in the data. The reference distribution is generated using kernel density estimation and a Gaussian copula framework. A dimension reduction strategy and sparse covariance estimation optimize this method for the high-dimensional, low-sample size setting. This method is similar to them method described in Helgeson and Bair (2016) <arXiv:1610.01424> except a Gaussian copula approach is used to account for feature correlation.
| Version: | 1.1.0 |
| Depends: | R (≥ 3.6.0) |
| Imports: | huge, PDSCE |
| Published: | 2020-04-13 |
| Author: | Erika S. Helgeson, David Vock, and Eric Bair |
| Maintainer: | Erika S. Helgeson <helge at umn.edu> |
| License: | GPL-2 | GPL-3 [expanded from: GPL (≥ 2)] |
| NeedsCompilation: | no |
| CRAN checks: | UNPaC results |
| Reference manual: | UNPaC.pdf |
| Package source: | UNPaC_1.1.0.tar.gz |
| Windows binaries: | r-devel: UNPaC_1.1.0.zip, r-release: UNPaC_1.1.0.zip, r-oldrel: UNPaC_1.1.0.zip |
| macOS binaries: | r-release: UNPaC_1.1.0.tgz, r-oldrel: UNPaC_1.1.0.tgz |
| Old sources: | UNPaC archive |
Please use the canonical form https://CRAN.R-project.org/package=UNPaC to link to this page.