Stationarity and invertibility of simple bilinear models

Necessary and sufficient conditions for strict stationarity and invertibility are found for one-parameter bilinear models. These conditions involve the expectations of the logarithms of the absolute values of the input and output sequences.     

On Drazin invertibility

The left Drazin spectrum and the Drazin spectrum coincide with the upper semi--Browder spectrum and the -Browder spectrum, respectively. We also prove that some spectra coincide whenever or satisfies the single-valued extension property.

     

The stationarity and invertibility of a class of nonlinear ARMA models

We investigate some probabilistic properties of a new class of nonlinear time series models. A sufficient condition for the existence of a unique causal, strictly and weakly stationary solution is derived. To understand the proposed models better, we further discuss the moment structure and obtain some Yule-Walker difference equations for the second and third order cumulants, which can also be used for identification purpose. A sufficient condition for invertibility is also provided.     

Ergodicity of observation-driven time series models and consistency of the maximum likelihood estimator

This paper deals with a general class of observation-driven time series models with a special focus on time series of counts. We provide conditions under which there exist strict-sense stationary and ergodic versions of such processes. The consistency of the maximum likelihood estimators is then derived for well-specified and misspecified models.     

On the estimation of nonlinear time series models

In the present paper, a framework for parametric estimation in nonlinear time series is developed. Strong consistency and asymptotic normality of minimum Hellinger distance estimates for a determined class of nonlinear models are investigated. The main Interest for these estimates is motivated by their robustness under perturbations as it has been emphazized in Beran [2]. The first part of the paper is devoted to the study of some probabilistic properties which ensure the existence and the optimal properties of the estimates     

Fractionally differenced models for water quality time series

This paper deals with the selection and evaluation of statistical techniques for use in the modeling and forecasting of water quality time series. The focus is on statistical concepts relevant to the analysis of flows and concentrations. A selection of time series procedures has been used for auditing water quality archival data, including the screening of data sets, correlation and spectrum calculations, and iterative model fitting. A summary is provided of experience with analyzing archival data on the Niagara River and the use of a fractionally differenced model.This paper is the result of a study performed for the International Joint Commission, United States and Canada. The authors gratefully acknowledge the direction and support provided by Dr. Joel L. Fisher.     

Forecasting time series with multiple seasonal patterns

A new approach is proposed for forecasting a time series with multiple seasonal patterns. A state space model is developed for the series using the innovations approach which enables us to develop explicit models for both additive and multiplicative seasonality. Parameter estimates may be obtained using methods from exponential smoothing. The proposed model is used to examine hourly and daily patterns in hourly data for both utility loads and traffic flows. Our formulation provides a model for several existing seasonal methods and also provides new options, which result in superior forecasting performance over a range of prediction horizons. In particular, seasonal components can be updated more frequently than once during a seasonal cycle. The approach is likely to be useful in a wide range of applications involving both high and low frequency data, and it handles missing values in a straightforward manner.     

一类随机环境下非线性时间序列模型的遍历性分析

本文主要讨论由模型Xn 1=h(en -q(n 1),-en,Xn 1-p(n 1),…Xn) en 1所确定的序列{Xn,n 1}的极限行为.     

Bilinear markovian representation and bilinear models

An extension of the linear Markovian repsentation called the bilinear Markovian representation is introduced, and is shown to provide representations of all-diagonal bilinear time series models. Some properties of the bilinear Markovian representation are also given.     

Application of time series analysis and modern control theory to the cement plant

Cement plant process is full of internal noise sources and feedback loops. Therefore, statistical approach is required to understand its dynamic characteristics. Time series analysis has been applied to some important subprocesses of a cement plant process. These are the vertical mill process, calcining process and clinker cooling process. Based on the AR models of these, a set of optimum controllers have been designed by modern control theory. Successful results of application are reported in this paper. A method of determining optimal production level is also discussed.     

Spline estimation of functional coefficient regression models for time series with correlated errors

In this work we focus on functional coefficient regression (FCR) models. Here we study the estimation of FCR models by splines, with autoregressive errors and show the rates of convergence of the proposed estimator. The importance of taking into account the correlation is assessed via simulation studies and multi-step ahead forecasts for a real data set.     

Latent class models for time series analysis

Latent class analysis of time series designed to classify and compare sets of series is discussed. For a particular time series in latent class the data are independently normally distributed with a vector of means, and common variance , that is, . The function of time, , can be represented by a linear combination of low-order splines (piecewise polynomials). The probability density function for the data of a time series is posited to be a finite mixture of spherical multivariate normal densities. The maximum-likelihood function is optimized by means of an EM algorithm. The stability of the estimates is investigated using a bootstrap procedure. Examples of real and artificial data are presented. Copyright © 1999 John Wiley & Sons, Ltd.     

A note on quadratic forms of stationary functional time series under mild conditions

We study distributional properties of a quadratic form of a stationary functional time series under mild moment conditions. As an important application, we obtain consistency rates of estimators of spectral density operators and prove joint weak convergence to a vector of complex Gaussian random operators. Weak convergence is established based on an approximation of the form via transforms of Hilbert-valued martingale difference sequences. As a side-result, the distributional properties of the long-run covariance operator are established.     

MODWT-ARMA model for time series prediction

Many time series in the applied sciences display a time-varying second order structure and long-range dependence (LRD). In this paper, we present a hybrid MODWT-ARMA model by combining the maximal overlap discrete wavelet transform (MODWT) and the ARMA model to deal with the non-stationary and LRD time series. We prove theoretically that the details series obtained by MODWT are stationary and short-range dependent (SRD). Then we derive the general form of MODWT-ARMA model. In the experimental study, the daily rainfall and Mackey–Glass time series are used to assess the performance of the hybrid model. Finally, the normalized error comparison with DWT-ARMA, EMD-ARMA and ARIMA model indicates that this combined model is an effective way to improve forecasting accuracy.     

On the invertibility of seasonally adjusted series

This paper examines the implications of the seasonal adjustment by an ARIMA model based (AMB) approach in the context of seasonal fractional integration. According to the AMB approach, if the model identified from the data contains seasonal unit roots, the adjusted series will not be invertible that has serious implications for the posterior analysis. We show that even if the ARIMA model identified from the data contains seasonal unit roots, if the true data generating process is stationary seasonally fractionally integrated (as it is often found in economic data), the AMB seasonal adjustment produces dips in the periodogram at seasonal frequencies, but the adjusted series still can be approximated by an invertible process. We also perform a small Monte Carlo study of the log-periodogram regression with tapered data for negative seasonal fractional integration. An empirical application for the Spanish economy that illustrates our results is also carried out at the end of the article.     

Spectral modeling of time series with missing data

Singular spectrum analysis is a natural generalization of principal component methods for time series data. In this paper we propose an imputation method to be used with singular spectrum-based techniques which is based on a weighted combination of the forecasts and hindcasts yield by the recurrent forecast method. Despite its ease of implementation, the obtained results suggest an overall good fit of our method, being able to yield a similar adjustment ability in comparison with the alternative method, according to some measures of predictive performance.     

A note on bilinear time series models

Conditions for the existence of a stationary solution for certain forms of bilinear difference equations are derived.     

Testing coefficients of AR and bilinear time series models by a graphical approach

AR and bilinear time series models are expressed as time series chain graphical models, based on which, it is shown that the coefficients of AR and bilinear models are the conditional correlation coefficients conditioned on the other components of the time series. Then a graphically based procedure is proposed to test the significance of the coefficients of AR and bilinear time series. Simulations show that our procedure performs well both in sizes and powers. This work was supported by the Hong Kong Polytechnic University Research Council, the National Natural Science Foundation of China (Grant No. 10671044) and the Science and Technology Bureau of Guangzhou Municipal Government of China (Grant No. LSBH-017)     

Testing coeffcients of AR and bilinear time series models by a graphical approach

AR and bilinear time series models are expressed as time series chain graphical models, based on which, it is shown that the coefficients of AR and bilinear models are the conditional correlation coefficients conditioned on the other components of the time series. Then a graphically based procedure is proposed to test the significance of the coeffcients of AR and bilinear time series. Simulations show that our procedure performs well both in sizes and powers.