Inference in Heavy-Tailed Nonstationary Multivariate Time Series

DOI10.1080/01621459.2022.2128807MaRDI QIDQ6154015zbMATH OpenOpenAlexFDO

Authors Matteo Barigozzi, Giuseppe Cavaliere, Lorenzo Trapani

Publication date 19 March 2024

Published in Journal of the American Statistical Association (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://arxiv.org/abs/2107.13894

zbMATH Keywords

factor modelsheavy tailsnonstationarityrandomized tests

Mathematics Subject Classification ID

Statistics (62-XX)

Abstract: We study inference on the common stochastic trends in a non-stationary, $N$-variate time series $y_t$, in the possible presence of heavy tails. We propose a novel methodology which does not require any knowledge or estimation of the tail index, or even knowledge as to whether certain moments (such as the variance) exist or not, and develop an estimator of the number of stochastic trends $m$ based on the eigenvalues of the sample second moment matrix of $y_t$. We study the rates of such eigenvalues, showing that the first $m$ ones diverge, as the sample size $T$ passes to infinity, at a rate faster by $Oleft(Tight)$ than the remaining $N-m$ ones, irrespective of the tail index. We thus exploit this eigen-gap by constructing, for each eigenvalue, a test statistic which diverges to positive infinity or drifts to zero according to whether the relevant eigenvalue belongs to the set of the first $m$ eigenvalues or not. We then construct a randomised statistic based on this, using it as part of a sequential testing procedure, ensuring consistency of the resulting estimator of $m$. We also discuss an estimator of the common trends based on principal components and show that, up to a an invertible linear transformation, such estimator is consistent in the sense that the estimation error is of smaller order than the trend itself. Finally, we also consider the case in which we relax the standard assumption of extit{i.i.d.} innovations, by allowing for heterogeneity of a very general form in the scale of the innovations. A Monte Carlo study shows that the proposed estimator for $m$ performs particularly well, even in samples of small size. We complete the paper by presenting four illustrative applications covering commodity prices, interest rates data, long run PPP and cryptocurrency markets.

Cites work

• A class of simple distribution-free rank-based unit root tests
• A modified information criterion for cointegration tests based on a VAR approximation
• Alternative asymptotics for cointegration tests in large VARs
• Bootstrap determination of the co-integration rank in vector autoregressive models
• Chover-type laws of the \(k\)-iterated logarithm for weighted sums of strongly mixing sequences
• Co-Integration and Error Correction: Representation, Estimation, and Testing
• Cointegrated processes with infinite variance innovations
• Cointegration in large VARs
• Determination of vector error correction models in high dimensions
• Eigenvalue ratio test for the number of factors
• Error-correction factor models for high-dimensional cointegrated time series
• Estimating cross-section common stochastic trends in nonstationary panel data
• Estimation and Hypothesis Testing of Cointegration Vectors in Gaussian Vector Autoregressive Models
• Estimation and Testing for Unit Roots in a Partially Nonstationary Vector Autoregressive Moving Average Model
• Estimation for Partially Nonstationary Multivariate Autoregressive Models
• Factor modeling for high-dimensional time series: inference for the number of factors
• Forecasting the term structure of government bond yields
• Heteroskedastic time series with a unit root
• scientific article; zbMATH DE number 614990 (Why is no real title available?)
• Identifying cointegration by eigenanalysis
• Inference in heavy-tailed vector error correction models
• Large covariance estimation through elliptical factor models
• Limit theory for moving averages of random variables with regularly varying tail probabilities
• Limit theory for the largest eigenvalues of sample covariance matrices with heavy-tails
• Limit theory for the sample covariance and correlation functions of moving averages
• Moments of Distributions Attracted to Stable Laws
• Non-stationary autoregressive processes with infinite variance
• Nonparametric cointegration analysis of fractional systems with unknown integration orders
• Nonparametric nonstationarity tests
• Nonstationary dynamic factor analysis
• Power-law distributions in empirical data
• Prediction and nonparametric estimation for time series with heavy tails
• Quantile factor models
• Robust factor number specification for large-dimensional elliptical factor model
• Self-Normalization for Time Series: A Review of Recent Developments
• Semiparametric error-correction models for cointegration with trends: pseudo-Gaussian and optimal rank-based tests of the cointegration rank
• Some mixing properties of time series models
• Spurious factor analysis
• Subsampling unit root tests for heavy-tailed observations
• Testing for (in)finite moments
• Testing for Common Trends
• Testing for randomness in a random coefficient autoregression model
• Testing Second-Order Dynamics for Autoregressive Processes in Presence of Time-Varying Variance
• Tests for cointegration with infinite variance errors
• TESTS OF COMMON STOCHASTIC TRENDS
• Time series regression on integrated continuous-time processes with heavy and light tails
• Unit root inference for non-stationary linear processes driven by infinite variance innovations
• Zipf's Law for Cities: An Explanation

Cited in

(4)

• Inference for the VEC(1) model with a heavy-tailed linear process errors*
• Online change-point detection for matrix-valued time series with latent two-way factor structure
• A new non-parametric Kendall's tau for matrix-valued elliptical observations
• Huber principal component analysis for large-dimensional factor models

This page was built for publication: Inference in Heavy-Tailed Nonstationary Multivariate Time Series

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6154015)