README.md

In SWotherspoon/bmixlm: Binary Mixtures of Linear Models

Binary Mixtures of Linear Models

Fit a binary mixture of linear models by Gibbs sampling. Assumes responses are drawn two subpopulations, each described by a different linear model, and it is unknown which responses are drawn from which subpopulation. The mixture model estimates the coefficient of the two subpopulation models and the probability that an observation is drawn from a particular subpopulation.

Installing

The package is easily installed from GitHub, using the devtools package.

devtools::install_github("SWotherspoon/bmixlm")

If you don't have devtools installed already, install it first.

install.packages("devtools")

(bmixlm otherwise does not need devtools for normal use.)

TODO

• Multiple Chains. Ideally the sampler would draw multiple chains. This would require rewriting the summary facilities to cope with multiple chains. This is a low priority.

• Parallelisation. At this point, the sampler is only capable of utilizing a single core on a multicore machine. It would be relatively simple to introduce coarse grain parallelism by having the sampler draw multiple chains in parallel, using something like the multicore facility in the parallel package. Unfortunately, at the time of writing there does not seem to be a good parallization solution that works equally well on all platforms. This is a low priority.

SWotherspoon/bmixlm documentation built on May 9, 2019, 12:07 p.m.