Giovanni Montana @ Imperial College London

PTM - Penalised mixtures of t distributions

 

We provide an R script for fitting the PTM model, a penalised finite mixture of Student's t distributions for robust model-based clustering. Gene selection and ranking is carried out by a data re-sampling approach.

Download

PTM.R - Example code

PTMFun.R - Main functions

 

 

 

References

Cozzini A., Jasra A. and Montana G. (2012) Model-based clustering with gene ranking using penalised mixtures of heavy-tailed distributions. Preprint