Test for association between the observed data and their systematic patterns of variations, that are often extracted by unsupervised learning. Systematic patterns may be captured by latent variables using principal component analysis (PCA), factor analysis (FA), and related methods. This allows one to, for example, obtain principal components (PCs) and conduct rigorous statistical testing for association between observed variables and PCs. Similarly, unsupervised clustering, such as K-means clustering, partition around medoids (PAM), and other algorithms, finds subpopulations among the observed variables. The jackstraw test can estimate statistical significance of cluster membership, so that one can evaluate the strength of membership assignments. This package also includes several related methods to support statistical inference and probabilistic feature selection for unsupervised learning.
| Version: | 1.2 |
| Depends: | R (≥ 3.0.0) |
| Imports: | corpcor, cluster, ClusterR, qvalue, methods, lfa, stats |
| Suggests: | parallel, knitr, rmarkdown |
| Published: | 2018-08-07 |
| Author: | Neo Christopher Chung, John D. Storey, Wei Hao |
| Maintainer: | Neo Christopher Chung <nchchung at gmail.com> |
| BugReports: | https://github.com/ncchung/jackstraw/issues |
| License: | GPL-2 |
| URL: | https://github.com/ncchung/jackstraw |
| NeedsCompilation: | no |
| Materials: | README |
| CRAN checks: | jackstraw results |
| Reference manual: | jackstraw.pdf |
| Vignettes: |
jackstraw: Statistical Inference using Latent Variables |
| Package source: | jackstraw_1.2.tar.gz |
| Windows binaries: | r-devel: jackstraw_1.2.zip, r-release: jackstraw_1.2.zip, r-oldrel: jackstraw_1.2.zip |
| OS X binaries: | r-release: jackstraw_1.2.tgz, r-oldrel: jackstraw_1.2.tgz |
| Old sources: | jackstraw archive |
Please use the canonical form https://CRAN.R-project.org/package=jackstraw to link to this page.