Full Model Estimation for Non-Parametric Multivariate Finite Mixture Models

This paper addresses the problem of full model estimation for non-parametric finite mixture models. It presents an approach for selecting the number of components and the subset of discriminative variables (i.e., the subset of variables having different distributions among the mixture components). The proposed approach considers a discretization of each variable into B bins and a penalization of the resulting log-likelihood. Considering that the number of bins tends to infinity as the sample size tends to infinity, we prove that our estimator of the model (number of components and subset of relevant variables for clustering) is consistent under a suitable choice of the penalty term. Interest of our proposal is illustrated on simulated and benchmark data.

READ FULL TEXT

page 15

page 31

page 32

research
08/10/2019

Estimation of the Number of Components of Non-Parametric Multivariate Finite Mixture Models

We propose a novel estimator for the number of components (denoted by M)...
research
01/06/2023

Non-parametric Multi-Partitions Clustering

In the framework of model-based clustering, a model, called multi-partit...
research
09/13/2022

Addressing overfitting in spectral clustering via a non-parametric bootstrap

Finite mixture modelling is a popular method in the field of clustering ...
research
02/19/2020

On the expected number of components in a finite admixture model

In this paper we describe the growth rate of the expected number of comp...
research
02/19/2020

On the geometric properties of finite mixture models

In this paper we relate the geometry of extremal points to properties of...
research
03/09/2022

Concave likelihood-based regression with finite-support response variables

We propose likelihood-based methods for regression when the response var...
research
12/13/2022

A smoothed semiparametric likelihood for estimation of nonparametric finite mixture models with a copula-based dependence structure

In this manuscript, we consider a finite multivariate nonparametric mixt...

Please sign up or login with your details

Forgot password? Click here to reset