Autoregressive and moving average models for zero‐inflated count time series

DOI10.1111/STAN.12255arXiv2004.10732OpenAlexW3202113053MaRDI QIDQ6089375FDOQ6089375

Authors: Siuli Mukhopadhyay, Rashmi Tiwari 

Publication date: 14 December 2023

Published in: Statistica Neerlandica (Search for Journal in Brave)

Abstract: Zero inflation is a common nuisance while monitoring disease progression over time. This article proposes a new observation driven model for zero inflated and over-dispersed count time series. The counts given the past history of the process and available information on covariates is assumed to be distributed as a mixture of a Poisson distribution and a distribution degenerate at zero, with a time dependent mixing probability, $p{i}_t$. Since, count data usually suffers from overdispersion, a Gamma distribution is used to model the excess variation, resulting in a zero inflated Negative Binomial (NB) regression model with mean parameter $lambd{a}_t$. Linear predictors with auto regressive and moving average (ARMA) type terms, covariates, seasonality and trend are fitted to $lambd{a}_t$ and $p{i}_t$ through canonical link generalized linear models. Estimation is done using maximum likelihood aided by iterative algorithms, such as Newton Raphson (NR) and Expectation and Maximization (EM). Theoretical results on the consistency and asymptotic normality of the estimators are given. The proposed model is illustrated using in-depth simulation studies and a dengue data set.

Full work available at URL: https://arxiv.org/abs/2004.10732

zbMATH Keywords

model selectionpredictionEM algorithmoverdispersionnegative binomialmixture distribution

Mathematics Subject Classification ID

Applications of statistics (62Pxx) Parametric inference (62Fxx) Inference from stochastic processes (62Mxx)

Cites Work

• Title not available (Why is that?)
• Likelihood Ratio Tests for Model Selection and Non-Nested Hypotheses
• Dynamic Generalized Linear Models and Bayesian Forecasting
• Likelihood analysis of non-Gaussian measurement time series
• First‐order integer valued AR processes with zero inflated poisson innovations
• Prediction and classification of non-stationary categorical time series
• Ergodicity of observation-driven time series models and consistency of the maximum likelihood estimator
• Observation-driven models for Poisson counts
• Theory and inference for a class of nonlinear models with application to time series of counts
• Poisson autoregression
• Title not available (Why is that?)
• Markov Regression Models for Time Series: A Quasi-Likelihood Approach
• Zero-Inflated Poisson Regression, with an Application to Defects in Manufacturing
• Generalized Autoregressive Moving Average Models
• Partial Likelihood Inference For Time Series Following Generalized Linear Models
• Time Series Models Based on Generalized Linear Models: Some Further Results
• Multivariate statistical modelling based on generalized linear models.
• Markov chain Monte Carlo for dynamic generalised linear models
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Time series analysis by state space methods.
• Monte Carlo maximum likelihood estimation for non-Gaussian state space models
• Monte Carlo EM Estimation for Time Series Models Involving Counts
• Time Series Analysis of Non-Gaussian Observations Based on State Space Models from Both Classical and Bayesian Perspectives
• Log-linear Poisson autoregression
• Dynamic Generalized Linear Models with Application to Environmental Epidemiology
• Estimation of the Negative Binomial Parameter κ by Maximum Quasi -Likelihood
• A NON‐GAUSSIAN FAMILY OF STATE‐SPACE MODELS WITH EXACT MARGINAL LIKELIHOOD
• Zero-inflated Poisson and negative binomial integer-valued GARCH models
• Zero-inflated and overdispersed: what's one to do?
• On the law of large numbers for (geometrically) ergodic Markov chains
• Penalized likelihood estimation and iterative Kalman smoothing for non-Gaussian dynamic regression models
• Markov regression models for count time series with excess zeros: a partial likelihood approach
• Time series regression for zero-inflated and overdispersed count data: a functional response model approach
• Modeling time series of count with excess zeros and ones based on INAR(1) model with zero-and-one inflated Poisson innovations
• Generalized ARMA models with martingale difference errors
• Decomposition of time series models in state-space form
• Adaptive log-linear zero-inflated generalized Poisson autoregressive model with applications to crime counts
• Stochastic Limit Theory
• COUNT AND DURATION TIME SERIES WITH EQUAL CONDITIONAL STOCHASTIC AND MEAN ORDERS

Cited In (3)

• Modeling and inference for counts time series based on zero-inflated exponential family INGARCH models
• Time series regression for zero-inflated and overdispersed count data: a functional response model approach
• Markov zero-inflated Poisson regression models for a time series of counts with excess zeros

This page was built for publication: Autoregressive and moving average models for zero‐inflated count time series

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6089375)