Model Averaging of Integer-Valued Autoregressive Model With Covariates
33 Pages Posted: 18 May 2021 Last revised: 5 Jan 2022
Date Written: May 15, 2021
Abstract
The integer-valued autoregressive (INAR) process is a class of structural models that can be used to model dependent count data in various fields including medicine, statistics, economics, finance and marketing. This paper proposes a K-fold cross-validation model averaging (KCVMA) method to average predictions from INAR models based on maximum likelihood estimation. The KCVMA method is shown to be asymptotically optimal in the sense of achieving the lowest quadratic loss. The KCVMA estimators are consistent, provided at least one candidate model is not underfitted. Monte Carlo simulations and empirical analysis illustrate the merits of the propose method relative to the existing model averaging and model selection methods
Keywords: Asymptotic optimality; K-fold cross-validation; Integer-valued autoregressive models; Maximum likelihood; Model averaging.
Suggested Citation: Suggested Citation