Multi-distributional Discriminant Analysis using Generalised Linear Latent Variable Modelling
genDA is a Discriminant Analysis (DA) algorithm capable for use in multi-distributional response data - generalising the capabilities of DA beyond Gaussian response. It utilises Generalised Linear Latent Variable Models (GLLVMs) to capture the covariance structure between the different response types and provide an efficient classifier for such datasets. This package leverages the highly resourceful TMB package for fast and accurate gradient calculation in C++.
This package is part of a suite of discriminant analysis packages we have authored for large-scale/complex datasets. See also our package multiDA, a statistical ML method for high dimensional, Gaussian data, with feature selection.
This work was presented at ACEMS - Enabling Algorithms Symposium in June, 2019. See a run down of the mathematics behind this package here.
# Install the development version from GitHub:
# install.packages("devtools")
devtools::install_github("sarahromanes/genDA")
- Sarah Romanes - @sarah_romanes
- John Ormerod - @john_t_ormerod
This project is licensed under the GPL-2 license.
- I am grateful to everyone who has provided thoughtful and helpful comments to support me in this project - especially Mark Greenaway for C++ implementation!