High-frequency sampling and kernel estimation for continuous-time moving average processes

DOI10.1111/JTSA.12022MaRDI QIDQ2852599zbMATH OpenOpenAlexFDO

Authors Vincenzo Ferrazzano, Claudia Klüppelberg, Peter J. Brockwell

Publication date 9 October 2013

Published in Journal of Time Series Analysis (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1107.4468

zbMATH Keywords

CARMA processcontinuous-time moving average processFICARMA processhigh-frequency datakernel estimationregular variationspectral theoryturbulenceWold representation

Mathematics Subject Classification ID

Prediction theory (aspects of stochastic processes) (60G25) Nonparametric estimation (62G05) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Inference from stochastic processes and prediction (62M20)

Abstract: Interest in continuous-time processes has increased rapidly in recent years, largely because of high-frequency data available in many applications. We develop a method for estimating the kernel function $g$ of a second-order stationary L'evy-driven continuous-time moving average (CMA) process $Y$ based on observations of the discrete-time process $Y^{D}elta$ obtained by sampling $Y$ at $Delta,2Delta,\mathrm{...},nDelta$ for small $Delta$. We approximate $g$ by $g^{D}elta$ based on the Wold representation and prove its pointwise convergence to $g$ as $Deltaightarrow0$ for $CARMA(p,q)$ processes. Two non-parametric estimators of $g^{D}elta$, based on the innovations algorithm and the Durbin-Levinson algorithm, are proposed to estimate $g$. For a Gaussian CARMA process we give conditions on the sample size $n$ and the grid-spacing $Delta(n)$ under which the innovations estimator is consistent and asymptotically normal as $nightarrowinfty$. The estimators can be calculated from sampled observations of {it any} CMA process and simulations suggest that they perform well even outside the class of CARMA processes. We illustrate their performance for simulated data and apply them to the Brookhaven turbulent wind speed data. Finally we extend results of citet{bfk:2011:1} for sampled CARMA processes to a much wider class of CMA processes.

Recommendations

• Noise recovery for Lévy-driven CARMA processes and high-frequency behaviour of approximating Riemann sums
• Spectral estimates for high-frequency sampled continuous-time autoregressive moving average processes
• High-frequency sampling of a continuous-time ARMA process
• Low-frequency estimation of continuous-time moving average Lévy processes
• Nonparametric estimation of the kernel function of symmetric stable moving average random functions

Cites work

• Abelian and Tauberian theorems relating the local behavior of an integrable function to the tail behavior of its Fourier transform
• Estimation of parameters of a continuous time Gaussian stationary process with rational spectral density
• Existence and uniqueness of stationary Lévy-driven CARMA processes
• Fourier analysis of time series. An introduction.
• Fractional Lévy processes with an application to long memory moving average processes
• High-frequency sampling of a continuous-time ARMA process
• scientific article; zbMATH DE number 2199141 (Why is no real title available?)
• scientific article; zbMATH DE number 3273551 (Why is no real title available?)
• scientific article; zbMATH DE number 3028224 (Why is no real title available?)
• Multipower variation for Brownian semistationary processes
• Noise recovery for Lévy-driven CARMA processes and high-frequency behaviour of approximating Riemann sums
• Progress in the Statistical Theory of Turbulence
• Simple consistent estimation of the coefficients of a linear filter
• The construction and estimation of continuous time models and discrete approximations in econometrics
• The Elementary Gaussian Processes
• THE ESTIMATION OF PARAMETERS IN SYSTEMS OF STOCHASTIC DIFFERENTIAL EQUATIONS
• Time series: theory and methods.
• Turbulent Flows

Cited in

(24)

• Stable continuous-time autoregressive process driven by stable subordinator
• Subordinated continuous-time AR processes and their application to modeling behavior of mechanical system
• Low-frequency estimation of continuous-time moving average Lévy processes
• Asymptotic moving average representation of high-frequency sampled multivariate CARMA processes
• Noise recovery for Lévy-driven CARMA processes and high-frequency behaviour of approximating Riemann sums
• Nonparametric estimation for i.i.d. Gaussian continuous time moving average models
• Nonparametric estimation of the kernel function of symmetric stable moving average random functions
• Kernel estimation for Lévy driven stochastic convolutions
• Lévy driven CARMA generalized processes and stochastic partial differential equations
• Lévy-driven causal CARMA random fields
• Dependence estimation for high-frequency sampled multivariate CARMA models
• High-frequency sampling of a continuous-time ARMA process
• Modelling turbulent time series by BSS-processes
• Limit theory for high frequency sampled MCARMA models
• Recent results in the theory and applications of CARMA processes
• scientific article; zbMATH DE number 7387569 (Why is no real title available?)
• On Lévy semistationary processes with a gamma kernel
• Stationary Gaussian Markov processes as limits of stationary autoregressive time series
• Selfdecomposable fields
• Limit theorems for the sample mean and sample autocovariances of continuous time moving averages driven by heavy-tailed Lévy noise
• Sampling, embedding and inference for CARMA processes
• Spectral estimates for high-frequency sampled continuous-time autoregressive moving average processes
• Semi-Lévy-driven CARMA process: estimation and prediction
• On the class of distributions of subordinated Lévy processes and bases

This page was built for publication: High-frequency sampling and kernel estimation for continuous-time moving average processes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2852599)