In fchamroukhi/mixRHLP: Mixture of Regression Models with Hidden Logistic Processes ('mixRHLP') for Simultaneous Curve Clustering and Segmentation

library(knitr)
knitr::opts_chunk$set(
    fig.align = "center",
    fig.height = 5.5,
    fig.width = 6,
    warning = FALSE,
    collapse = TRUE,
    dev.args = list(pointsize = 10),
    out.width = "90%",
    par = TRUE
)
knit_hooks$set(par = function(before, options, envir)
  { if (before && options$fig.show != "none") 
       par(family = "sans", mar = c(4.1,4.1,1.1,1.1), mgp = c(3,1,0), tcl = -0.5)
})

library(mixRHLP)

Introduction

mixRHLP is a R package for segmentation. It provides functions for parameter estimation via the EM algorithm. This document gives a quick tour of mixRHLP (version r packageVersion("mixRHLP")) functionalities. It was written in R Markdown, using the knitr package for production. See help(package="mixRHLP") for further details and references provided by citation("mixRHLP").

Load data

data("toydataset")
x <- toydataset$x
Y <- t(toydataset[,2:ncol(toydataset)])

Set up mixRHLP model parameters

K <- 3 # Number of clusters
R <- 3 # Number of regimes (polynomial regression components)
p <- 1 # Degree of the polynomials
q <- 1 # Order of the logistic regression (by default 1 for contiguous segmentation)
variance_type <- "heteroskedastic" # "heteroskedastic" or "homoskedastic" model

Set up EM parameters

n_tries <- 1
max_iter <- 1000
threshold <- 1e-5
verbose <- TRUE
verbose_IRLS <- FALSE
init_kmeans <- TRUE

Estimation

mixrhlp <- emMixRHLP(X = x, Y = Y, K, R, p, q, variance_type, init_kmeans, 
                     n_tries, max_iter, threshold, verbose, verbose_IRLS)

Summary

mixrhlp$summary()

Plots

Estimated signal

mixrhlp$plot(what = "estimatedsignal")

Regressors

mixrhlp$plot(what = "regressors")

Log-likelihood

mixrhlp$plot(what = "loglikelihood")

fchamroukhi/mixRHLP documentation built on Sept. 23, 2019, 4:19 a.m.