An empirical likelihood approach for symmetric α-stable processes

Fumiya Akashi, Yan Liu, Masanobu Taniguchi

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Empirical likelihood approach is one of non-parametric statistical methods, which is applied to the hypothesis testing or construction of confidence regions for pivotal unknown quantities. This method has been applied to the case of independent identically distributed random variables and second order stationary processes. In recent years, we observe heavy-tailed data in many fields. To model such data suitably, we consider symmetric scalar and multivariate á-stable linear processes generated by infinite variance innovation sequence. We use a Whittle likelihood type estimating function in the empirical likelihood ratio function and derive the asymptotic distribution of the empirical likelihood ratio statistic for á-stable linear processes. With the empirical likelihood statistic approach, the theory of estimation and testing for second order stationary processes is nicely extended to heavy-tailed data analyses, not straightforward, and applicable to a lot of financial statistical analyses.

Original languageEnglish
Pages (from-to)2093-2119
Number of pages27
JournalBernoulli
Volume21
Issue number4
DOIs
Publication statusPublished - 2015 Nov 1

Fingerprint

Symmetric Stable Processes
Empirical Likelihood
Linear Process
Stable Process
Stationary Process
Whittle Likelihood
Infinite Variance
Estimating Function
Likelihood Ratio Statistic
Confidence Region
Nonparametric Methods
Likelihood Ratio
Hypothesis Testing
Identically distributed
Statistical method
Data Model
Asymptotic distribution
Statistic
Random variable
Scalar

Keywords

  • Confidence region
  • Empirical likelihood ratio
  • Heavy tail
  • Normalized power transfer function
  • Self-normalized periodogram
  • Symmetric α-stable process
  • Whittle likelihood

ASJC Scopus subject areas

  • Statistics and Probability

Cite this

An empirical likelihood approach for symmetric α-stable processes. / Akashi, Fumiya; Liu, Yan; Taniguchi, Masanobu.

In: Bernoulli, Vol. 21, No. 4, 01.11.2015, p. 2093-2119.

Research output: Contribution to journalArticle

Akashi, Fumiya ; Liu, Yan ; Taniguchi, Masanobu. / An empirical likelihood approach for symmetric α-stable processes. In: Bernoulli. 2015 ; Vol. 21, No. 4. pp. 2093-2119.
@article{4b174276afc343b48cf47c1fefb41aaa,
title = "An empirical likelihood approach for symmetric α-stable processes",
abstract = "Empirical likelihood approach is one of non-parametric statistical methods, which is applied to the hypothesis testing or construction of confidence regions for pivotal unknown quantities. This method has been applied to the case of independent identically distributed random variables and second order stationary processes. In recent years, we observe heavy-tailed data in many fields. To model such data suitably, we consider symmetric scalar and multivariate {\'a}-stable linear processes generated by infinite variance innovation sequence. We use a Whittle likelihood type estimating function in the empirical likelihood ratio function and derive the asymptotic distribution of the empirical likelihood ratio statistic for {\'a}-stable linear processes. With the empirical likelihood statistic approach, the theory of estimation and testing for second order stationary processes is nicely extended to heavy-tailed data analyses, not straightforward, and applicable to a lot of financial statistical analyses.",
keywords = "Confidence region, Empirical likelihood ratio, Heavy tail, Normalized power transfer function, Self-normalized periodogram, Symmetric α-stable process, Whittle likelihood",
author = "Fumiya Akashi and Yan Liu and Masanobu Taniguchi",
year = "2015",
month = "11",
day = "1",
doi = "10.3150/14-BEJ636",
language = "English",
volume = "21",
pages = "2093--2119",
journal = "Bernoulli",
issn = "1350-7265",
publisher = "International Statistical Institute",
number = "4",

}

TY - JOUR

T1 - An empirical likelihood approach for symmetric α-stable processes

AU - Akashi, Fumiya

AU - Liu, Yan

AU - Taniguchi, Masanobu

PY - 2015/11/1

Y1 - 2015/11/1

N2 - Empirical likelihood approach is one of non-parametric statistical methods, which is applied to the hypothesis testing or construction of confidence regions for pivotal unknown quantities. This method has been applied to the case of independent identically distributed random variables and second order stationary processes. In recent years, we observe heavy-tailed data in many fields. To model such data suitably, we consider symmetric scalar and multivariate á-stable linear processes generated by infinite variance innovation sequence. We use a Whittle likelihood type estimating function in the empirical likelihood ratio function and derive the asymptotic distribution of the empirical likelihood ratio statistic for á-stable linear processes. With the empirical likelihood statistic approach, the theory of estimation and testing for second order stationary processes is nicely extended to heavy-tailed data analyses, not straightforward, and applicable to a lot of financial statistical analyses.

AB - Empirical likelihood approach is one of non-parametric statistical methods, which is applied to the hypothesis testing or construction of confidence regions for pivotal unknown quantities. This method has been applied to the case of independent identically distributed random variables and second order stationary processes. In recent years, we observe heavy-tailed data in many fields. To model such data suitably, we consider symmetric scalar and multivariate á-stable linear processes generated by infinite variance innovation sequence. We use a Whittle likelihood type estimating function in the empirical likelihood ratio function and derive the asymptotic distribution of the empirical likelihood ratio statistic for á-stable linear processes. With the empirical likelihood statistic approach, the theory of estimation and testing for second order stationary processes is nicely extended to heavy-tailed data analyses, not straightforward, and applicable to a lot of financial statistical analyses.

KW - Confidence region

KW - Empirical likelihood ratio

KW - Heavy tail

KW - Normalized power transfer function

KW - Self-normalized periodogram

KW - Symmetric α-stable process

KW - Whittle likelihood

UR - http://www.scopus.com/inward/record.url?scp=84941955764&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84941955764&partnerID=8YFLogxK

U2 - 10.3150/14-BEJ636

DO - 10.3150/14-BEJ636

M3 - Article

AN - SCOPUS:84941955764

VL - 21

SP - 2093

EP - 2119

JO - Bernoulli

JF - Bernoulli

SN - 1350-7265

IS - 4

ER -