On diagnostics in conditionally heteroskedastic time series models under elliptical distributions

Research output: Contribution to journalArticle

20 Citations (Scopus)

Abstract

In statistical diagnostics and sensitivity analysis, the local influence method plays an important rôle. In the present paper, we use this method to study financial time series data and conditionally heteroskedastic models under elliptical distributions. We start with a likelihood displacement, and consider data- and model-perturbation schemes. We obtain corresponding matrices of derivatives, and measures of slope and normal curvature, and then discuss the assessment of local influence.

Original languageEnglish
Pages (from-to)393-405
Number of pages13
JournalJournal of Applied Probability
Volume41A
DOIs
Publication statusPublished - 1 Jan 2004
Externally publishedYes

Fingerprint

Elliptical Distribution
Local Influence
Time Series Models
Diagnostics
Likelihood Displacement
Normal Curvature
Financial Data
Financial Time Series
Time Series Data
Sensitivity Analysis
Slope
Perturbation
Derivative
Model
Elliptical distribution
Time series models
Financial time series
Curvature
Time series data
Derivatives

Cite this

@article{fc003f5fb2e74d859704b520138fd933,
title = "On diagnostics in conditionally heteroskedastic time series models under elliptical distributions",
abstract = "In statistical diagnostics and sensitivity analysis, the local influence method plays an important r{\^o}le. In the present paper, we use this method to study financial time series data and conditionally heteroskedastic models under elliptical distributions. We start with a likelihood displacement, and consider data- and model-perturbation schemes. We obtain corresponding matrices of derivatives, and measures of slope and normal curvature, and then discuss the assessment of local influence.",
keywords = "Likelihood displacement, Matrices of derivatives, Matrix differential, Observed information matrix, Statistical diagnostics",
author = "Shuangzhe Liu",
year = "2004",
month = "1",
day = "1",
doi = "10.1239/jap/1082552214",
language = "English",
volume = "41A",
pages = "393--405",
journal = "Journal of Applied Probability",
issn = "0021-9002",
publisher = "University of Sheffield",

}

On diagnostics in conditionally heteroskedastic time series models under elliptical distributions. / Liu, Shuangzhe.

In: Journal of Applied Probability, Vol. 41A, 01.01.2004, p. 393-405.

Research output: Contribution to journalArticle

TY - JOUR

T1 - On diagnostics in conditionally heteroskedastic time series models under elliptical distributions

AU - Liu, Shuangzhe

PY - 2004/1/1

Y1 - 2004/1/1

N2 - In statistical diagnostics and sensitivity analysis, the local influence method plays an important rôle. In the present paper, we use this method to study financial time series data and conditionally heteroskedastic models under elliptical distributions. We start with a likelihood displacement, and consider data- and model-perturbation schemes. We obtain corresponding matrices of derivatives, and measures of slope and normal curvature, and then discuss the assessment of local influence.

AB - In statistical diagnostics and sensitivity analysis, the local influence method plays an important rôle. In the present paper, we use this method to study financial time series data and conditionally heteroskedastic models under elliptical distributions. We start with a likelihood displacement, and consider data- and model-perturbation schemes. We obtain corresponding matrices of derivatives, and measures of slope and normal curvature, and then discuss the assessment of local influence.

KW - Likelihood displacement

KW - Matrices of derivatives

KW - Matrix differential

KW - Observed information matrix

KW - Statistical diagnostics

UR - http://www.scopus.com/inward/record.url?scp=33845777957&partnerID=8YFLogxK

U2 - 10.1239/jap/1082552214

DO - 10.1239/jap/1082552214

M3 - Article

VL - 41A

SP - 393

EP - 405

JO - Journal of Applied Probability

JF - Journal of Applied Probability

SN - 0021-9002

ER -