Journal of Time Series Analysis最新文献

{"title":"Estimation on unevenly spaced time series","authors":"Liudas Giraitis,&nbsp;Fulvia Marotta","doi":"10.1111/jtsa.12704","DOIUrl":"10.1111/jtsa.12704","url":null,"abstract":"<p>In many different fields realizations of stationary time series might be recorded at irregular points in time, resulting in observed unevenly spaced samples. These missing observations can happen for several reasons, depending on the mechanisms that record the data or external conditions that force the missing observations. In this article, we first focus on the question if we can estimate the mean of a stationary time series when data are not equally spaced. We show that any unevenly spaced sample can be used to estimate the mean of an underlying stationary linear time series. Specifically, we do not impose any restrictions on sampling structure and times, as long as they are independent of the underlying time series. We provide an expression for the sample mean estimator and we establish its asymptotic properties and the central limit theorem. Subsequently we studentize estimation which allows to build confidence intervals for the mean. Finite sample properties of the estimator for the mean are investigated in a Monte Carlo study which confirms good performance of such estimation procedure.</p>","PeriodicalId":49973,"journal":{"name":"Journal of Time Series Analysis","volume":"44 5-6","pages":"556-577"},"PeriodicalIF":0.9,"publicationDate":"2023-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42297921","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Weighted l1-Penalized Corrected Quantile Regression for High-Dimensional Temporally Dependent Measurement Errors","authors":"Monika Bhattacharjee,&nbsp;Nilanjan Chakraborty,&nbsp;Hira L. Koul","doi":"10.1111/jtsa.12703","DOIUrl":"10.1111/jtsa.12703","url":null,"abstract":"<p>This article derives some large sample properties of weighted <math>\u0000 <mrow>\u0000 <msub>\u0000 <mrow>\u0000 <mi>l</mi>\u0000 </mrow>\u0000 <mrow>\u0000 <mn>1</mn>\u0000 </mrow>\u0000 </msub>\u0000 </mrow></math>-penalized corrected quantile estimators of the regression parameter vector in a high-dimensional errors in variables (EIVs) linear regression model. In this model, the number of predictors <math>\u0000 <mrow>\u0000 <mi>p</mi>\u0000 </mrow></math> depends on the sample size <math>\u0000 <mrow>\u0000 <mi>n</mi>\u0000 </mrow></math> and tends to infinity, generally at a faster rate than <math>\u0000 <mrow>\u0000 <mi>n</mi>\u0000 </mrow></math>, as <math>\u0000 <mrow>\u0000 <mi>n</mi>\u0000 </mrow></math> tends to infinity. Moreover, the measurement errors in the covariates are assumed to have linear stationary temporal dependence and known Laplace marginal distribution while the regression errors are assumed to be independent non-identically distributed random variables having possibly heavy tails. The article discusses some rates of consistency of these estimators, a model consistency result and an appropriate data adaptive algorithm for obtaining a suitable choice of weights. A simulation study assesses the finite sample performance of some of the proposed estimators. This article also contains analogs of Massart's inequality for independent and short memory moving average predictors, which is instrumental in establishing the said consistency rates of the above mentioned estimators in the current setup of high dimensional EIVs regression models.</p>","PeriodicalId":49973,"journal":{"name":"Journal of Time Series Analysis","volume":"44 5-6","pages":"442-473"},"PeriodicalIF":0.9,"publicationDate":"2023-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/jtsa.12703","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42440178","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Testing for symmetric correlation matrices with applications to factor models","authors":"Nan-Jung Hsu,&nbsp;Lai Heng Sim,&nbsp;Ruey S. Tsay","doi":"10.1111/jtsa.12702","DOIUrl":"10.1111/jtsa.12702","url":null,"abstract":"Factor models have been widely used in recent years to model high‐dimensional spatio‐temporal data. However, the validity of employing factor models in a specific application has received less attention. This article proposes test statistics for testing the symmetry in cross‐correlation matrices of a high‐dimensional stochastic process implied by exact factor models. A rejection of symmetry indicates that the use of an exact factor model is questionable. Both simulations and real examples are used to demonstrate the applications and to study the finite‐sample performance of the proposed test statistics. Empirical results show that the proposed test statistics are effective in identifying cases where exact factor models are not appropriate, providing valuable guidance for choosing factor models in a high‐dimensional setting.","PeriodicalId":49973,"journal":{"name":"Journal of Time Series Analysis","volume":"44 5-6","pages":"622-643"},"PeriodicalIF":0.9,"publicationDate":"2023-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/jtsa.12702","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45524669","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Nonlinear kernel mode-based regression for dependent data","authors":"Tao Wang","doi":"10.1111/jtsa.12700","DOIUrl":"10.1111/jtsa.12700","url":null,"abstract":"<p>Under stationary <math>\u0000 <mrow>\u0000 <mi>α</mi>\u0000 </mrow></math>-mixing dependent samples, we in this article develop a novel nonlinear regression based on mode value for time series sequences to achieve robustness without sacrificing estimation efficiency. The estimation process is built on a kernel-based objective function with a constant bandwidth (tuning parameter) that is independent of sample size and can be adjusted to maximize efficiency. The asymptotic distribution of the resultant estimator is established under suitable conditions, and the convergence rate is demonstrated to be the same as that in nonlinear mean regression. To numerically estimate the kernel mode-based regression, we develop a modified modal-expectation-maximization algorithm in conjunction with Taylor expansion. A robust Wald-type test statistic derived from the resulting estimator is also provided, along with its asymptotic distribution for the null and alternative hypotheses. The local robustness of the proposed estimation procedure is studied using influence function analysis, and the good finite sample performance of the newly suggested model is verified through Monte Carlo simulations. We finally combine the recommended kernel mode-based regression with neural networks to develop a kernel mode-based neural networks model, the performance of which is evidenced by an empirical examination of exchange rate prediction.</p>","PeriodicalId":49973,"journal":{"name":"Journal of Time Series Analysis","volume":"45 2","pages":"189-213"},"PeriodicalIF":0.9,"publicationDate":"2023-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42733193","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Correcting the bias of the sample cross-covariance estimator","authors":"Yifan Li","doi":"10.1111/jtsa.12701","DOIUrl":"10.1111/jtsa.12701","url":null,"abstract":"<p>We derive the finite sample bias of the sample cross-covariance estimator based on a stationary vector-valued time series with an unknown mean. This result leads to a bias-corrected estimator of cross-covariances constructed from linear combinations of sample cross-covariances, which can in theory correct for the bias introduced by the first <math>\u0000 <mrow>\u0000 <mi>h</mi>\u0000 </mrow></math> lags of cross-covariance with any <math>\u0000 <mrow>\u0000 <mi>h</mi>\u0000 </mrow></math> not larger than the sample size minus two. Based on the bias-corrected cross-covariance estimator, we propose a bias-adjusted long run covariance (LRCOV) estimator. We derive the asymptotic relations between the bias-corrected estimators and their conventional Counterparts in both the small-<math>\u0000 <mrow>\u0000 <mi>b</mi>\u0000 </mrow></math> and the fixed-<math>\u0000 <mrow>\u0000 <mi>b</mi>\u0000 </mrow></math> limit. Simulation results show that: (i) our bias-corrected cross-covariance estimators are very effective in eliminating the finite sample bias of their conventional counterparts, with potential mean squared error reduction when the data generating process is highly persistent; and (ii) the bias-adjusted LRCOV estimators can have superior performance to their conventional counterparts with a smaller bias and mean squared error.</p>","PeriodicalId":49973,"journal":{"name":"Journal of Time Series Analysis","volume":"45 2","pages":"214-247"},"PeriodicalIF":0.9,"publicationDate":"2023-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135792157","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Announcement: Call for Papers for Special Issue in Honour of Stephen J. Taylor","authors":"Torben Andersen,&nbsp;Kim Christensen,&nbsp;Ingmar Nolte","doi":"10.1111/jtsa.12693","DOIUrl":"10.1111/jtsa.12693","url":null,"abstract":"","PeriodicalId":49973,"journal":{"name":"Journal of Time Series Analysis","volume":"44 4","pages":"336"},"PeriodicalIF":0.9,"publicationDate":"2023-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45926665","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Corrigendum: Error bounds and asymptotic expansions for Toeplitz product functionals of unbounded spectra","authors":"Tetsuya Takabatake","doi":"10.1111/jtsa.12690","DOIUrl":"10.1111/jtsa.12690","url":null,"abstract":"<p>We investigate error orders for integral limit approximations to traces of products of Toeplitz matrices generated by integrable functions on <math>\u0000 <mrow>\u0000 <mo>[</mo>\u0000 <mo>−</mo>\u0000 <mi>π</mi>\u0000 <mo>,</mo>\u0000 <mi>π</mi>\u0000 <mo>]</mo>\u0000 </mrow></math> having some singularities at the origin. Even though a sharp error order of the above approximation is derived in Theorem 2 of Lieberman and Phillips (2004, <i>Journal of Time Series Analysis</i>, 25(5) 733–753), its proof contains an inaccuracy. In the present article, we reinvestigate error orders of the above trace approximation problem and rigorously validate the sharp error order derived in Lieberman and Phillips (2004, <i>Journal of Time Series Analysis</i>, 25(5) 733–753).</p>","PeriodicalId":49973,"journal":{"name":"Journal of Time Series Analysis","volume":"45 1","pages":"158-160"},"PeriodicalIF":0.9,"publicationDate":"2023-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/jtsa.12690","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43489912","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Stochastic local and moderate departures from a unit root and its application to unit root testing","authors":"Mikihito Nishi,&nbsp;Eiji Kurozumi","doi":"10.1111/jtsa.12691","DOIUrl":"10.1111/jtsa.12691","url":null,"abstract":"<p>Local-to-unity and moderate-deviations specifications have been popular alternatives to unit root modeling. This article considers another kind of departures from a unit root, of the form <math>\u0000 <mrow>\u0000 <mi>c</mi>\u0000 <msub>\u0000 <mrow>\u0000 <mi>v</mi>\u0000 </mrow>\u0000 <mrow>\u0000 <mi>t</mi>\u0000 </mrow>\u0000 </msub>\u0000 <mo>/</mo>\u0000 <msup>\u0000 <mrow>\u0000 <mi>T</mi>\u0000 </mrow>\u0000 <mrow>\u0000 <mi>β</mi>\u0000 </mrow>\u0000 </msup>\u0000 </mrow></math>, where <math>\u0000 <mrow>\u0000 <msub>\u0000 <mrow>\u0000 <mi>v</mi>\u0000 </mrow>\u0000 <mrow>\u0000 <mi>t</mi>\u0000 </mrow>\u0000 </msub>\u0000 </mrow></math> is random and <math>\u0000 <mrow>\u0000 <mi>β</mi>\u0000 </mrow></math> determines the distance from a unit root. We classify the stochastic departures into two types: local and moderate. This classification task is completed by investigating the asymptotic behavior of unit root tests that assume the stochastic unit root (STUR) processes as the alternative hypothesis. The stochastic local-to-unity model arises when <math>\u0000 <mrow>\u0000 <mi>β</mi>\u0000 <mo>=</mo>\u0000 <mn>3</mn>\u0000 <mo>/</mo>\u0000 <mn>4</mn>\u0000 </mrow></math>; in this case, the test statistics have limiting distributions different from those under the unit root null, and their asymptotic powers are greater than size. Moderate deviations emerge when <math>\u0000 <mrow>\u0000 <mn>1</mn>\u0000 <mo>/</mo>\u0000 <mn>2</mn>\u0000 <mo>≤</mo>\u0000 <mi>β</mi>\u0000 <mo>&lt;</mo>\u0000 <mn>3</mn>\u0000 <mo>/</mo>\u0000 <mn>4</mn>\u0000 </mrow></math>, in which case the test statistics diverge. We also propose new tests for a unit root against an STUR, whose construction is based on the limit theory developed in this article. To evaluate the performance of these new tests, we derive the limiting Gaussian power envelope under the local alternative from an approximate model.</p>","PeriodicalId":49973,"journal":{"name":"Journal of Time Series Analysis","volume":"45 1","pages":"133-157"},"PeriodicalIF":0.9,"publicationDate":"2023-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/jtsa.12691","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44475299","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Portmanteau tests for periodic ARMA models with dependent errors","authors":"Y. Boubacar Maïnassara,&nbsp;A. Ilmi Amir","doi":"10.1111/jtsa.12692","DOIUrl":"10.1111/jtsa.12692","url":null,"abstract":"<p>In this article, we derive the asymptotic distributions of residual and normalized residual empirical autocovariances and autocorrelations of (parsimonious) periodic autoregressive moving-average (PARMA) models under the assumption that the errors are uncorrelated but not necessarily independent. We then deduce the modified portmanteau statistics. We establish the asymptotic behavior of the proposed statistics. It is shown that the asymptotic distribution of the modified portmanteau tests is that of a weighted sum of independent chi-squared random variables, which can be different from the usual chi-squared approximation used under independent and identically distributed assumption on the noise. We also propose another test based on a self-normalization approach to check the adequacy of PARMA models. A set of Monte Carlo experiments and an application to financial data are presented.</p>","PeriodicalId":49973,"journal":{"name":"Journal of Time Series Analysis","volume":"45 2","pages":"164-188"},"PeriodicalIF":0.9,"publicationDate":"2023-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"64272733","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Clustering multivariate time series using energy distance","authors":"Richard A. Davis,&nbsp;Leon Fernandes,&nbsp;Konstantinos Fokianos","doi":"10.1111/jtsa.12688","DOIUrl":"10.1111/jtsa.12688","url":null,"abstract":"<p>A novel methodology is proposed for clustering multivariate time series data using energy distance defined in Székely and Rizzo (2013). Specifically, a dissimilarity matrix is formed using the energy distance statistic to measure the separation between the finite-dimensional distributions for the component time series. Once the pairwise dissimilarity matrix is calculated, a hierarchical clustering method is then applied to obtain the dendrogram. This procedure is completely nonparametric as the dissimilarities between stationary distributions are directly calculated without making any model assumptions. In order to justify this procedure, asymptotic properties of the energy distance estimates are derived for general stationary and ergodic time series. The method is illustrated in a simulation study for various component time series that are either linear or nonlinear. Finally, the methodology is applied to two examples; one involves the GDP of selected countries and the other is the population size of various states in the U.S.A. in the years 1900–1999.</p>","PeriodicalId":49973,"journal":{"name":"Journal of Time Series Analysis","volume":"44 5-6","pages":"487-504"},"PeriodicalIF":0.9,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44254149","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}