A Non-Iterative Quantile Change Detection Method in Mixture Model with Heavy-Tailed Components

06/19/2020
by   Yuantong Li, et al.
0

Estimating parameters of mixture model has wide applications ranging from classification problems to estimating of complex distributions. Most of the current literature on estimating the parameters of the mixture densities are based on iterative Expectation Maximization (EM) type algorithms which require the use of either taking expectations over the latent label variables or generating samples from the conditional distribution of such latent labels using the Bayes rule. Moreover, when the number of components is unknown, the problem becomes computationally more demanding due to well-known label switching issues <cit.>. In this paper, we propose a robust and quick approach based on change-point methods to determine the number of mixture components that works for almost any location-scale families even when the components are heavy tailed (e.g., Cauchy). We present several numerical illustrations by comparing our method with some of popular methods available in the literature using simulated data and real case studies. The proposed method is shown be as much as 500 times faster than some of the competing methods and are also shown to be more accurate in estimating the mixture distributions by goodness-of-fit tests.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
01/24/2017

Robust mixture modelling using sub-Gaussian stable distribution

Heavy-tailed distributions are widely used in robust mixture modelling d...
research
06/22/2015

Non-Normal Mixtures of Experts

Mixture of Experts (MoE) is a popular framework for modeling heterogenei...
research
03/13/2021

Extreme value analysis for mixture models with heavy-tailed impurity

This paper deals with the extreme value analysis for the triangular arra...
research
04/08/2020

Robust Mixture Modeling using Weighted Complete Estimating Equations

Mixture modeling that takes account of potential heterogeneity in data i...
research
11/21/2014

Robust EM kernel-based methods for linear system identification

Recent developments in system identification have brought attention to r...
research
12/02/2022

Compound Batch Normalization for Long-tailed Image Classification

Significant progress has been made in learning image classification neur...
research
10/17/2014

Inference and Mixture Modeling with the Elliptical Gamma Distribution

We study modeling and inference with the Elliptical Gamma Distribution (...

Please sign up or login with your details

Forgot password? Click here to reset