Software

Code for the methods developed.

• MixCircReg

  Fits mixtures of circular regressions with circular and linear covariates using the EM algorithm.

  Written in R, accompanying Model-based clustering using a new mixture of circular regressions.

• SPGMNRs

  Fits semi-parametric Gaussian mixtures of non-parametric regressions with a model-based ECM algorithm, alongside the local EM and naive EM algorithms for comparison.

  Written in R, accompanying A modified EM-type algorithm to estimate semi-parametric mixtures of non-parametric regressions.

• NPCGMRs

  Code and data for nonparametric contaminated Gaussian mixtures of regressions.

  Written in R, accompanying Nonparametric contaminated Gaussian mixture of regressions.

• S-CG-MoLE

  Fits robust semi-parametric mixtures of linear experts with contaminated Gaussian errors.

  Written in R, accompanying Robust semi-parametric mixtures of linear experts using the contaminated Gaussian distribution.