Finite mixture model selection for total energy consumption

Abstract

To estimate the amount of total energy consumption to dissension of energy production for future, estimating the statistical model is necessary. The goal of this article is using Vuong's test for non-nested and misspecified finite mixture models, where at the same time we consider the model selection criteria. The simulation study confirms using Vuong's test for mixture models. Based on theoretical results we verified some model selection criteria and Test to selection a model for energy consumption data from Washington, Oregon and California sities in USA. Our data was acquired from http://www.eia.gov/.

Model Selection for Mixture Models Using Perfect Sample

“…He has proposed two types of coefficients for the mixture criterion, one based on the density and another one based on the risk function. Fallahigilan and Sayyareh (2016) have selected model for mixture distribution with the Vuong's test, AIC and BIC, based on simulation study. Wichitchan et al (2018) apply the idea of goodness of fit (GOF) testing procedure to finite mixture models and investigate their performance when performing hypothesis testing of mixture models for different types of alternative hypotheses.…”