Integer-valued Time Series Research Articles

Parameters Estimation in Non-Negative Integer-Valued Time Series: Approach Based on Probability Generating Functions

This manuscript deals with a parameter estimation of a non-negative integer-valued (NNIV) time series based on the so-called probability generating function (PGF) method. The theoretical background of the PGF estimation technique for a very general, stationary class of NNIV time series is described, as well as the asymptotic properties of the obtained estimates. After that, a particular emphasis is given to PGF estimators of independent identical distributed (IID) and integer-valued non-negative autoregressive (INAR) series. A Monte Carlo study of the thus obtained PGF estimates, based on a numerical integration of the appropriate objective function, is also presented. For this purpose, numerical quadrature formulas were computed using Gegenbauer orthogonal polynomials. Finally, the application of the PGF estimators in the dynamic analysis of some actual data is given.

Test for Conditional Variance of Integer-Valued Time Series

Test for Conditional Variance of Integer-Valued Time Series

A NEGATIVE BINOMIAL AUTOREGRESSION WITH A LINEAR CONDITIONAL VARIANCE-TO-MEAN FUNCTION

A general integer-valued time-series model with a conditional variance proportional to the conditional mean is proposed. Specifically, the conditional distribution is a Poisson mixture with a dependent mixing sequence, which results in a negative binomial distribution with a linear conditional variance-to-mean relationship. In addition, the conditional mean is specified as a general parametric function of past observations. We first propose stationarity, ergodicity, and finite moment conditions for the model. Furthermore, the parameters are estimated using the Poisson quasi-maximum likelihood estimate, whose asymptotic properties are studied under weak conditions. Illustrations of the proposed methodology on simulated and actual time series of counts are given.

Statistical modelling of COVID-19 and drug data via an INAR(1) process with a recent thinning operator and cosine Poisson innovations.

In this paper, we propose the first-order stationary integer-valued autoregressive process with the cosine Poisson innovation, based on the negative binomial thinning operator. It can be equi-dispersed, under-dispersed and over-dispersed. Therefore, it is flexible for modelling integer-valued time series. Some statistical properties of the process are derived. The parameters of the process are estimated by two methods of estimation and the performances of the estimators are evaluated via some simulation studies. Finally, we demonstrate the usefulness of the proposed model by modelling and analyzing some practical count time series data on the daily deaths of COVID-19 and the drug calls data.

Review of Copula for Bivariate Distributions of Zero-Inflated Count Time Series Data

The class of bivariate integer-valued time series models, described via copula theory, is gaining popularity in the literature because of applications in health sciences, engineering, financial management and more. Each time series follows a Markov chain with the serial dependence captured using copula-based distribution functions from the Poisson and the zero-inflated Poisson margins. The copula theory is again used to capture the dependence between the two series.
 
 However, the efficiency and adaptability of the copula are being challenged because of the discrete nature of data and also in the case of zero-inflation of count time series. Likelihood-based inference is used to estimate the model parameters for simulated and real data with the bivariate integral of copula functions. While such copula functions offer great flexibility in capturing dependence, there remain challenges related to identifying the best copula type for a given application.  This paper presents a survey of the literature on bivariate copula for discrete data with an emphasis on the zero-inflated nature of the modelling. We demonstrate additional experiments on to confirm that the copula has potential as greater research area.

A New Bivariate INAR(1) Model with Time-Dependent Innovation Vectors

Recently, there has been a growing interest in integer-valued time series models, especially in multivariate models. Motivated by the diversity of the infinite-patch metapopulation models, we propose an extension to the popular bivariate INAR(1) model, whose innovation vector is assumed to be time-dependent in the sense that the mean of the innovation vector is linearly increased by the previous population size. We discuss the stationarity and ergodicity of the observed process and its subprocesses. We consider the conditional maximum likelihood estimate of the parameters of interest, and establish their large-sample properties. The finite sample performance of the estimator is assessed via simulations. Applications on crime data illustrate the model.

Coherent Forecasting for a Mixed Integer-Valued Time Series Model

In commerce, economics, engineering and the sciences, quantitative methods based on statistical models for forecasting are very useful tools for prediction and decision. There is an abundance of papers on forecasting for continuous-time series but relatively fewer papers for time series of counts which require special consideration due to the integer nature of the data. A popular method for modelling is the method of mixtures which is known for its flexibility and thus improved prediction capability. This paper studies the coherent forecasting for a flexible stationary mixture of Pegram and thinning (MPT) process, and develops the likelihood-based asymptotic distribution. Score functions and the Fisher information matrix are presented. Numerical studies are used to assess the performance of the forecasting methods. Also, a comparison is made with existing discrete-valued time series models. Finally, the practical application is illustrated with two sets of real data. It is shown that the mixture model provides good forecasting performance.

Re-visiting the COVID-19 analysis using the class of high ordered integer-valued time series models with harmonic features

The COVID-19 series is obviously one of the most volatile time series with lots of spikes and oscillations. The conventional integer-valued auto-regressive time series models (INAR) may be limited to account for such features in COVID-19 series such as severe over-dispersion, excess of zeros, periodicity, harmonic shapes and oscillations. This paper proposes alternative formulations of the classical INAR process by considering the class of high-ordered INAR models with harmonic innovation distributions. Interestingly, the paper further explores the bivariate extension of these high-ordered INARs. South Africa and Mauritius’ COVID-19 series are re-scrutinized under the optic of these new INAR processes. Some simulation experiments are also executed to validate the new models and their estimation procedures.

Penalized empirical likelihood inference for the GINAR(p) model

Integer-valued time series data are greatly useful in many applications, such as reporting the daily number of patients impacted by an epidemic. When modelling high-order integer-valued time series data, order selection is a difficult task. In this paper, we propose a penalized empirical likelihood (PEL) method for order selection and parameter estimation in generalized pth-order integer-valued autoregressive (GINAR(p)) model. We show that the PEL method in the GINAR(p) model has oracle properties, which means that the PEL estimators identify the model as efficiently as if the true structure of the model was known ahead of time. Furthermore, we present the PEL ratio statistic to test a linear hypothesis of the parameter and demonstrate that it has an asymptotically distribution under the null hypothesis. Numerical simulation and real data analysis are carried out to assess the performance our proposed method.

Asymptotic negative binomial quasi-likelihood inference for periodic integer-valued time series models

This paper establishes the strong consistency and asymptotic normality of the negative binomial quasi-maximum likelihood estimator (NBQMLE) for a general class of integer-valued time series models whose parameters are periodic time-varying. This class of models is specified via its conditional mean, which is expressed as a periodic parametric function of (infinite) past observations. An illustration on specific models, namely the Poisson periodic INGARCH model, the negative binomial periodic INGARCH model, and the periodic INAR(1) model is given. In addition, the performance of the NBQMLE is assessed through a simulation study, and an application of the proposed methodology to a real dataset is provided.

A nonparametric Bayesian analysis for meningococcal disease counts based on integer-valued threshold time series models

To better describe the characteristics of time series of counts showing over-dispersion, asymmetry and structural change features, this paper considers a class of random coefficient integer-valued threshold autoregressive processes that properly capture flexible asymmetric and nonlinear responses without assuming the distributions for the errors. By the Taylor expansion, a normally distributed working likelihood is obtained and the Bayesian empirical likelihood inference is conducted for the model parameters. Through the normalized approximation of the nonparametric likelihood, the originally complicated Bayesian calculation has been greatly simplified. Finally, the proposed method is verified via some simulations and an empirical analysis of the meningococcal disease data set in Germany.

Marginal likelihood estimation for the negative binomial INGARCH model

In recent years, there has been increased interest in modeling integer-valued time series. Many methods for time series of counts have been developed in the literature because of their wide applications to epidemiology, finance, disease modeling and environmental science. The negative binomial integer-valued generalized autoregressive conditional heteroscedasticity model is a popular one, which can deal with both over-dispersion and potential extreme observations. The accurate estimation of the parameters in the model is extremely important. We adopt the marginal likelihood to estimate the intercept parameter and maximum likelihood to estimate other parameters of the model. We conduct simulations to assess the performance of this estimation method, and compare it with that of estimating all model parameters by maximum likelihood. The results show the superiority of proposed estimation method. We use two real examples to illustrate the model’s ability to fit over-dispersed data and the validity of the estimation method.

Bayesian modelling of integer-valued transfer function models

External events are commonly known as interventions that often affect times series of counts. This research introduces a class of transfer function models that include four different types of interventions on integer-valued time series: abrupt start and abrupt decay (additive outlier), abrupt start and gradual decay (transient shift), abrupt start and permanent effect (level shift) and gradual start and permanent effect. We propose integer-valued transfer function models incorporating a generalized Poisson, log-linear generalized Poisson or negative binomial to estimate and detect these four types of interventions in a time series of counts. Utilizing Bayesian methods, which are adaptive Markov chain Monte Carlo (MCMC) algorithms to obtain the estimation, we further employ deviance information criterion (DIC), posterior odd ratios and mean squared standardized residual for model comparisons. As an illustration, this study evaluates the effectiveness of our methods through a simulation study and application to crime data in Albury City, New South Wales (NSW) Australia. Simulation results show that the MCMC procedure is reasonably effective. The empirical outcome also reveals that the proposed models are able to successfully detect the locations and type of interventions.

Cluster point processes and Poisson thinning INARMA

In this paper, we consider Poisson thinning Integer-valued time series models, namely integer-valued moving average model (INMA) and Integer-valued Autoregressive Moving Average model (INARMA), and their relationship with cluster point processes, the Cox point process and the dynamic contagion process. We derive the probability generating functionals of INARMA models and compare to that of cluster point processes. The main aim of this paper is to prove that, under a specific parametric setting, INMA and INARMA models are just discrete versions of continuous cluster point processes and hence converge weakly when the length of subintervals goes to zero.

Re-analyzing the SARS-CoV-2 series using an extended integer-valued time series models: A situational assessment of the COVID-19 in Mauritius.

This paper proposes some high-ordered integer-valued auto-regressive time series process of order p (INAR(p)) with Zero-Inflated and Poisson-mixtures innovation distributions, wherein the predictor functions in these mentioned distributions allow for covariate specification, in particular, time-dependent covariates. The proposed time series structures are tested suitable to model the SARs-CoV-2 series in Mauritius which demonstrates excess zeros and hence significant over-dispersion with non-stationary trend. In addition, the INAR models allow the assessment of possible causes of COVID-19 in Mauritius. The results illustrate that the event of Vaccination and COVID-19 Stringency index are the most influential factors that can reduce the locally acquired COVID-19 cases and ultimately, the associated death cases. Moreover, the INAR(7) with Zero-inflated Negative Binomial innovations provides the best fitting and reliable Root Mean Square Errors, based on some short term forecasts. Undeniably, these information will hugely be useful to Mauritian authorities for implementation of comprehensive policies.

Inference for nonstationary time series of counts with application to change-point problems

We consider an integer-valued time series \((Y_t)_{t\in {\mathbb {Z}}}\) where the model after a time \(k^*\) is Poisson autoregressive with the conditional mean that depends on a parameter \(\theta ^*\in \varTheta \subset {\mathbb {R}}^d\). The structure of the process before \(k^*\) is unknown; it could be any other integer-valued process, that is, \((Y_t)_{t\in {\mathbb {Z}}}\) could be nonstationary. It is established that the maximum likelihood estimator of \(\theta ^*\) computed on the nonstationary observations is consistent and asymptotically normal. Subsequently, we carry out the sequential change-point detection in a large class of Poisson autoregressive models, and propose a monitoring scheme for detecting change. The procedure is based on an updated estimator, which is computed without the historical observations. The above results of inference in a nonstationary setting are applied to prove the consistency of the proposed procedure. A simulation study as well as a real data application are provided.

On MCMC sampling in self-exciting integer-valued threshold time series models

Markov Chain Monte Carlo (MCMC) methods have been shown to be a useful tool in many branches in statistics. However, due to the complex structure of the models, this method remains an open problem for threshold integer-valued time series models. This study develops Bayesian inference for a class of self-exciting integer-valued threshold autoregressive models, which is implemented by means of a new MCMC algorithm. By introducing the latent variables series, a complete data likelihood is obtained. Based on which, the full conditional distributions are easily obtained with familiar forms. Furthermore, by maximizing the complete data likelihood, the threshold parameter is also accurately estimated. Finally, the performance of the MCMC algorithm is evaluated via some simulations and a real data example.

Temporal aggregation and systematic sampling for INGARCH processes

The temporal aggregation (TA) and systematic sampling (SS) in continuous-valued time series were widely studied, while the TA and SS in integer-valued time series obtain very little attention. In this paper, we thoroughly discuss them based on the integer-valued generalized autoregressive conditional heteroskedasticity (INGARCH) process. On one hand, the TA and SS for INGARCH processes are studied, and related predictors are discussed. We also illustrate that the TA or SS provides approaches to estimate INGARCH processes with different frequencies. On the other hand, definitions of the strong and weak INGARCH processes are given. The aggregated INGARCH and sampled INGARCH processes are shown to be weak, that is, they satisfy a weak characterization of INGARCH processes in terms of linear projection. Quasi maximum likelihood and nonlinear least squares estimation methods are considered for the weak INGARCH processes. Related asymptotic properties and simulation results are discussed. Finally, an empirical example illustrates the applicability of our results.

The ARMA Point Process and its Estimation

An autoregressive–moving-average (ARMA) point process model is introduced, which combines self-exciting and shot-noise cluster mechanisms, both useful in a variety of applications. The process is analogous to the ARMA for integer-valued time series, sharing methodological and mathematical similarities. A maximum likelihood estimation procedure, based on MCEM (Monte Carlo Expectation Maximization), is derived and studied. This approach conveniently allows for: (i) trends in immigration/background intensity, (ii) multiple parametric specifications of memory functions and mark distributions, as well as (iii) cases where marks and immigrants are not observed. As such, the ARMA point process provides a flexible framework to disentangle cluster mechanisms in continuously observed count data.

First-Order Random Coefficient Multinomial Autoregressive Model for Finite-Range Time Series of Counts

In view of the complexity and asymmetry of finite range multi-state integer-valued time series data, we propose a first-order random coefficient multinomial autoregressive model in this paper. Basic probabilistic and statistical properties of the model are discussed. Conditional least squares (CLS) and weighted conditional least squares (WCLS) estimators of the model parameters are derived, and their asymptotic properties are established. In simulation studies, we compare these two methods with the conditional maximum likelihood (CML) method to verify the proposed procedure. A real example is applied to illustrate the advantages of our model.