Discrete-valued Time Series Research Articles

Several Approaches for the Prediction of the Operating Modes of a Wind Turbine

Growing concern about climate change has intensified efforts to use renewable energy, with wind energy highlighted as a growing source. It is known that wind turbines are characterized by distinct operating modes that reflect production efficiency. In this work, we focus on the forecasting problem for univariate discrete-valued time series of operating modes. We define three prediction strategies to overcome the difficulties associated with missing data. These strategies are evaluated through experiments using five forecasting methods across two real-life datasets. Two of the forecasting methods have been introduced in the statistical literature as extensions of the well-known context algorithm: variable length Markov chains and Bayesian context tree. Additionally, we consider a Bayesian method based on conditional tensor factorization and two different smoothers from the classical tools for time series forecasting. After evaluating each pair prediction strategy/forecasting method in terms of prediction accuracy versus computational complexity, we provide guidance on the methods that are suitable for forecasting the time series of operating modes. The prediction results that we report demonstrate that high accuracy can be achieved with reduced computational resources.

Generalized ordinal patterns in discrete-valued time series: nonparametric testing for serial dependence

We provide a new testing procedure to detect serial dependence in time series. Our method is based solely on the ordinal structure of the data. We explicitly allow for ties in the data windows we consider. Consequently, we use generalised ordinal patterns, that is, Cayley permutations. Unlike in the classical case, the pattern distribution is not uniform under the null hypothesis of serial independence. In our new framework, the underlying distribution has to be taken into account and we overcome this problem by a bootstrap procedure. The applicability of our method is supported by a simulation study and two real-world data examples.

Strong mixing properties of discrete-valued time series with exogenous covariates

We derive strong mixing conditions for many existing discrete-valued time series models that include exogenous covariates in the dynamic. Our main contribution is to study how a mixing condition on the covariate process transfers to a mixing condition for the response. Using a coupling method, we first derive mixing conditions for some Markov chains in random environments, which gives a first result for some autoregressive categorical processes with strictly exogenous regressors. Our result is then extended to some infinite memory categorical processes. In the second part of the paper, we study autoregressive models for which the covariates are sequentially exogenous. Using a general random mapping approach on finite sets, we get explicit mixing conditions that can be checked for many categorical time series found in the literature, including multinomial autoregressive processes, ordinal time series and dynamic multiple choice models. We also study some autoregressive count time series using a somewhat different contraction argument. Our contribution fills an important gap for such models, presented here under a more general form, since such a strong mixing condition is often assumed in some recent works but no general approach is available to check it.

Two Features of the GINAR(1) Process and Their Impact on the Run-Length Performance of Geometric Control Charts.

The geometric first-order integer-valued autoregressive process (GINAR(1)) can be particularly useful to model relevant discrete-valued time series, namely in statistical process control. We resort to stochastic ordering to prove that the GINAR(1) process is a discrete-time Markov chain governed by a totally positive order 2 (TP2) transition matrix.Stochastic ordering is also used to compare transition matrices referring to pairs of GINAR(1) processes with different values of the marginal mean. We assess and illustrate the implications of these two stochastic ordering results, namely on the properties of the run length of geometric charts for monitoring GINAR(1) counts.

Epidemic change-point detection in general integer-valued time series

ABSTRACT In this paper, we consider the structural change in a class of discrete valued time series, where the true conditional distribution of the observations is assumed to be unknown. The conditional mean of the process depends on a parameter θ ∗ which may change over time. We provide sufficient conditions for the consistency and the asymptotic normality of the Poisson quasi-maximum likelihood estimator (QMLE) of the model. We consider an epidemic change-point detection and propose a test statistic based on the QMLE of the parameter. Under the null hypothesis of a constant parameter (no change), the test statistic converges to a distribution obtained from increments of a Browninan bridge. The test statistic diverges to infinity under the epidemic alternative, which establishes that the proposed procedure is consistent in power. The effectiveness of the proposed procedure is illustrated by simulated and real data examples.

Adaptation of Partial Mutual Information from Mixed Embedding to Discrete-Valued Time Series.

A causality analysis aims at estimating the interactions of the observed variables and subsequently the connectivity structure of the observed dynamical system or stochastic process. The partial mutual information from mixed embedding (PMIME) is found appropriate for the causality analysis of continuous-valued time series, even of high dimension, as it applies a dimension reduction by selecting the most relevant lag variables of all the observed variables to the response, using conditional mutual information (CMI). The presence of lag components of the driving variable in this vector implies a direct causal (driving-response) effect. In this study, the PMIME is appropriately adapted to discrete-valued multivariate time series, called the discrete PMIME (DPMIME). An appropriate estimation of the discrete probability distributions and CMI for discrete variables is implemented in the DPMIME. Further, the asymptotic distribution of the estimated CMI is derived, allowing for a parametric significance test for the CMI in the DPMIME, whereas for the PMIME, there is no parametric test for the CMI and the test is performed using resampling. Monte Carlo simulations are performed using different generating systems of discrete-valued time series. The simulation suggests that the parametric significance test for the CMI in the progressive algorithm of the DPMIME is compared favorably to the corresponding resampling significance test, and the accuracy of the DPMIME in the estimation of direct causality converges with the time-series length to the accuracy of the PMIME. Further, the DPMIME is used to investigate whether the global financial crisis has an effect on the causality network of the financial world market.

Bivariate Poisson 2Sum-Lindley Distributions and the Associated BINAR(1) Processes

Discrete-valued time series modeling has witnessed numerous bivariate first-order integer-valued autoregressive process or BINAR(1) processes based on binomial thinning and different innovation distributions. These BINAR(1) processes are mainly focused on over-dispersion. This paper aims to propose new bivariate distributions and processes based on a recently proposed over-dispersed distribution: the Poisson 2S-Lindley distribution. The new bivariate distributions, denoted by the abbreviations BP2S-L(I) and BP2S-L(II), are then used as innovation distributions for the BINAR(1) process. Properties are investigated for both distributions as well as for the BINAR(1) processes. The distribution parameters are estimated using the maximum likelihood method, and the BINAR(1)BP2S-L(I) and BINAR(1)BP2S-L(II) process parameters are estimated using the conditional least squares and conditional maximum likelihood methods. Monte Carlo simulation experiments are conducted to study large and small sample performances and for the comparison of the estimation methods. The Pittsburgh crime series and candy sales datasets are then used to compare the new BINAR(1) processes to some other existing BINAR(1) processes in the literature.

Discrete-valued time series based on the exponential family with the multidimensional parameter and their probabilistic and statistical analysis.

We propose herein a new parsimonious Markov model for a discrete-valued time series with conditional probability distributions of observations lying in the exponential family with the multidimensional parameter. A family of explicit consistent asymptotically normal statistical estimators is constructed for the parameters of the proposed model for increasing length of observed time series, and asymptotically effective estimator is found within this constructed family. The obtained results can be used for robust statistical analysis of discrete-valued time series,and for statistical analysis of discrete-valued spatio-temporal data and random fields.

Interaction networks from discrete event data by Poisson multivariate mutual information estimation and information flow with applications from gene expression data

In this work, we introduce a new methodology for inferring the interaction structure of discrete valued time series which are Poisson distributed. While most related methods are premised on continuous state stochastic processes, in fact, discrete and counting event oriented stochastic process are natural and common, so called time-point processes. An important application that we focus on here is gene expression, where it is often assumed that the data is generated from a multivariate Poisson distribution. Nonparameteric methods such as the popular k-nearest neighbors are slow converging for discrete processes, and thus data hungry. Now, with the new multi-variate Poisson estimator developed here as the core computational engine, the causation entropy (CSE) principle, together with the associated greedy search algorithm optimal CSE (oCSE) allows us to efficiently infer the true network structure for this class of stochastic processes that were previously not practical. We illustrate the power of our method, first in benchmarking with synthetic datum, and then by inferring the genetic factors network from a breast cancer micro-ribonucleic acid sequence count data set. We show the Poisson oCSE gives the best performance among the tested methods and discovers previously known interactions on the breast cancer data set.

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.

Variable selection in sparse GLARMA models

In this paper, we propose a novel and efficient two-stage variable selection approach for sparse GLARMA models, which are pervasive for modelling discrete-valued time series. Our approach consists in iteratively combining the estimation of the autoregressive moving average (ARMA) coefficients of GLARMA models with regularized methods designed for performing variable selection in regression coefficients of Generalized Linear Models (GLM). We first establish the consistency of the ARMA part coefficient estimators in a specific case. Then, we explain how to efficiently implement our approach. Finally, we assess the performance of our methodology using synthetic data, compare it with alternative methods and illustrate it on an example of real-world application. Our approach, which is implemented in the GlarmaVarSel R package, is very attractive since it benefits from a low computational load and is able to outperform the other methods in terms of recovering the non-null regression coefficients.

Omen: discovering sequential patterns with reliable prediction delays

Suppose we are given a discrete-valued time series X of observed events and an equally long binary sequence Y that indicates whether something of interest happened at that particular point in time. We consider the problem of mining serial episodes, sequential patterns allowing for gaps, from X that reliably predict those interesting events. With reliable we mean patterns that not only predict that an interesting event is likely to follow, but in particular that we can also accurately tell how how long until that event will happen. In other words, we are specifically interested in patterns with a highly skewed distribution of delays between pattern occurrences and predicted events. As it is unlikely that a single pattern can explain a complex real-world progress, we are after the smallest, least redundant set of such patterns that together explain the interesting events well. We formally define this problem in terms of the Minimum Description Length principle, by which we identify the best patterns as those that describe the occurrences of interesting events Y most succinctly given the data over X . As neither discovering the optimal explanation of Y given a set of patterns, nor the discovery of optimal pattern set are problems that allow for straightforward optimization, we break the problem in two and propose effective heuristics for both. Through extensive empirical evaluation, we show that both our main method, Omen, and its fast approximation fOmen, work well in practice and both quantitatively and qualitatively beat the state of the art.

Observation-driven models for discrete-valued time series

Statistical inference for discrete-valued time series has not been developed like traditional methods for time series generated by continuous random variables. Some relevant models exist, but the lack of a homogenous framework raises some critical issues. For instance, it is not trivial to explore whether models are nested and it is quite arduous to derive stochastic properties which simultaneously hold across different specifications. In this paper, inference for a general class of first order observation-driven models for discrete-valued processes is developed. Stochastic properties such as stationarity and ergodicity are derived under easy-to-check conditions, which can be directly applied to all the models encompassed in the class and for every distribution which satisfies mild moment conditions. Consistency and asymptotic normality of quasi-maximum likelihood estimators are established, with the focus on the exponential family. Finite sample properties and the use of information criteria for model selection are investigated throughout Monte Carlo studies. An empirical application to count data is discussed, concerning a test-bed time series on the spread of an infection.

Neural network-based models of binomial time series in data analysis problems

This article is devoted to constructing neural network-based models for discrete-valued time series and their use in computer data analysis. A new family of binomial time series based on neural networks is presented, which makes it possible to approximate the arbitrary-type stochastic dependence in time series. Ergodicity conditions and an equivalence relation for these models are determined. Consistent statistical estimators for model parameters and algorithms for computer data analysis (including forecasting and pattern recognition) are developed.

On Bayesian model selection for INGARCH models viatrans-dimensional Markov chain Monte Carlo methods

There is an increasing interest in models for discrete valued time series. Among them, the integer autoregressive conditional heteroscedastic (INGARCH) is a model that has found several applications. In the present article, we study the problem of model selection for this family of models. Namely we consider that an observation conditional on the past follows a Poisson distribution where its mean depends on its past mean values and on past observations. We consider both linear and log-linear models. Our purpose is to select the most appropriate order of such models, using a trans-dimensional Bayesian approach that allows jumps between competing models. A small simulation experiment supports the usage of the method. We apply the methodology to real datasets to illustrate the potential of the approach.

Statistical analysis of multivariate discrete-valued time series

This work gives an overview of statistical analysis for some models for multivariate discrete-valued (MDV) time series. We present observation-driven models and models based on higher-order Markov chains. Several extensions are highlighted including non-stationarity, network autoregressions, conditional non-linear autoregressive models, robust estimation, random fields and spatio-temporal models.

On Abrupt Changes of the Approximate and Sample Entropy Values in Supercomputer Power Consumption

When calculating the Approximate Entropy (ApEn) and the Sample Entropy (SampEn) of a time series, representing the course of supercomputer power consumption, seemingly unexplained abrupt changes sometimes appear. These sudden changes occur when calculating the course of ApEn and SampEn values using a floating time window and can be manifested also in other discrete-valued time series, such as the development of prices on stock exchanges, time series from birth-death models, count-data time series, etc. These abrupt changes are not continuous and it is clear that they do not reflect real changes in the complexity degree of the analyzed time series. This can cause erroneous detection of a change in the degree of complexity of the time series, which can trigger a false alarm that something unusual is happening in the complex system being monitored (human brain, engine gearbox, financial market, supercomputer infrastructure, etc.). This work reveals in detail the mechanism of this phenomenon and also proposes measures to prevent its occurrence.

Clustering discrete-valued time series

There is a need for the development of models that are able to account for discreteness in data, along with its time series properties and correlation. Our focus falls on INteger-valued AutoRegressive (INAR) type models. The INAR type models can be used in conjunction with existing model-based clustering techniques to cluster discrete-valued time series data. With the use of a finite mixture model, several existing techniques such as the selection of the number of clusters, estimation using expectation-maximization and model selection are applicable. The proposed model is then demonstrated on real data to illustrate its clustering applications.

Statistical Analysis of Discrete-valued Time Series by Parsimonious High-order Markov Chains

Problems of statistical analysis of discrete-valued time series are considered. Two approaches for construction of parsimonious (small-parametric) models for observed discrete data are proposed based on high-order Markov chains.Consistent statistical estimators for parameters of the developed models and some known models, and also statistical tests on the values of parameters are constructed. Probabilistic properties of the constructed statistical inferences are given. The developed theory is also applied for statistical analysis of spatio-temporal data. Theoretical results are illustrated by computer experiments on real statistical data.

The family of the bivariate integer-valued autoregressive process (BINAR(1)) with Poisson–Lindley (PL) innovations

ABSTRACTIn the literature of discrete-valued time series modelling, various bivariate integer-valued autoregressive time series models of order 1 (BINAR(1)) have been proposed particularly based on the binomial thinning mechanism and with different innovation distributions. These BINAR(1)s are mostly suitable for modelling bivariate counting series of varied levels of overdispersion. Recently, in the context of overdispersion, the INAR(1) with Poisson–Lindley (PL) innovations have been introduced to provide superior model fitness criteria than other competing INAR(1)s. Thus, this paper proposes to develop classes of BINAR(1)PL, including BINAR(1)PL(I) and BINAR(1)PL(II), under different cross-correlation functions. The parameter estimations are conducted via the conditional maximum likelihood (CML) approach. Monte Carlo simulation experiments are implemented to assess the asymptotic properties of the CML estimators under different combinations of the cross-correlation parameters. The proposed models are also applied to the Pittsburg crimes data and compared with other competing popular overdispersed BINAR(1) models.