bslda
Sparse linear discriminant analysis using network-guided block covariance matrix (bslda), which allows researchers to utilize prior network information to guide structure of covariance matrix. ‘bslda’ package provides computationally efficient and user friendly interface to fit the bslda models.
library(bslda)
package?bslda
class?bslda
Installation
The stable versions of ‘bslda’ package can be obtained from the following URLs:
Package source: https://github.com/dongjunchung/chunglab_binary_packages/blob/master/bslda_0.1.0.tar.gz
Windows binary: https://github.com/dongjunchung/chunglab_binary_packages/blob/master/bslda_0.1.0.zip
Mac OS/X binary: comming soon
To install the developmental versions of ‘bslda’ package, it’s easiest to use the ‘devtools’ package.
#install.packages("devtools")
library(devtools)
install_github("elflini/bslda")
References
Nam JH, Kim D, and Chung D (2019) ‘‘Sparse linear discrimintn analysis using network-guided block covariance matrix’’.