Construction and visualization of confidence sets for frequentist distributional forecasts

David Harris, Gael M. Martin, Indeewara Perera, D. S. Poskitt

Research output: Contribution to journalArticleResearchpeer-review

Abstract

The focus of this article is on the quantification of sampling variation in frequentist probabilistic forecasts. We propose a method of constructing confidence sets that respects the functional nature of the forecast distribution, and use animated graphics to visualize the impact of parameter uncertainty on the location, dispersion, and shape of the distribution. The confidence sets are derived via the inversion of a Wald test, and the ellipsoid that defines the boundary of the set computed numerically. A wide range of linear and nonlinear time series models—encompassing long memory, state space, and mixture specifications—is used to demonstrate the procedure, based on artificially generated data. An empirical example in which distributional forecasts of both financial returns and its stochastic volatility are produced is then used to illustrate the practical importance of accommodating sampling variation in the manner proposed.

Original languageEnglish
Pages (from-to)92-104
Number of pages13
JournalJournal of Computational and Graphical Statistics
Volume28
Issue number1
DOIs
Publication statusPublished - 2019

Keywords

  • Animated graphics
  • Confidence regions
  • Heterogeneous autoregressive model
  • Probabilistic forecasts
  • Realized volatility
  • Time series models

Cite this

@article{ea1a4928c90042c18631fd1427aced21,
title = "Construction and visualization of confidence sets for frequentist distributional forecasts",
abstract = "The focus of this article is on the quantification of sampling variation in frequentist probabilistic forecasts. We propose a method of constructing confidence sets that respects the functional nature of the forecast distribution, and use animated graphics to visualize the impact of parameter uncertainty on the location, dispersion, and shape of the distribution. The confidence sets are derived via the inversion of a Wald test, and the ellipsoid that defines the boundary of the set computed numerically. A wide range of linear and nonlinear time series models—encompassing long memory, state space, and mixture specifications—is used to demonstrate the procedure, based on artificially generated data. An empirical example in which distributional forecasts of both financial returns and its stochastic volatility are produced is then used to illustrate the practical importance of accommodating sampling variation in the manner proposed.",
keywords = "Animated graphics, Confidence regions, Heterogeneous autoregressive model, Probabilistic forecasts, Realized volatility, Time series models",
author = "David Harris and Martin, {Gael M.} and Indeewara Perera and Poskitt, {D. S.}",
year = "2019",
doi = "10.1080/10618600.2018.1476252",
language = "English",
volume = "28",
pages = "92--104",
journal = "Journal of Computational and Graphical Statistics",
issn = "1061-8600",
publisher = "Taylor & Francis",
number = "1",

}

Construction and visualization of confidence sets for frequentist distributional forecasts. / Harris, David; Martin, Gael M.; Perera, Indeewara; Poskitt, D. S.

In: Journal of Computational and Graphical Statistics, Vol. 28, No. 1, 2019, p. 92-104.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Construction and visualization of confidence sets for frequentist distributional forecasts

AU - Harris, David

AU - Martin, Gael M.

AU - Perera, Indeewara

AU - Poskitt, D. S.

PY - 2019

Y1 - 2019

N2 - The focus of this article is on the quantification of sampling variation in frequentist probabilistic forecasts. We propose a method of constructing confidence sets that respects the functional nature of the forecast distribution, and use animated graphics to visualize the impact of parameter uncertainty on the location, dispersion, and shape of the distribution. The confidence sets are derived via the inversion of a Wald test, and the ellipsoid that defines the boundary of the set computed numerically. A wide range of linear and nonlinear time series models—encompassing long memory, state space, and mixture specifications—is used to demonstrate the procedure, based on artificially generated data. An empirical example in which distributional forecasts of both financial returns and its stochastic volatility are produced is then used to illustrate the practical importance of accommodating sampling variation in the manner proposed.

AB - The focus of this article is on the quantification of sampling variation in frequentist probabilistic forecasts. We propose a method of constructing confidence sets that respects the functional nature of the forecast distribution, and use animated graphics to visualize the impact of parameter uncertainty on the location, dispersion, and shape of the distribution. The confidence sets are derived via the inversion of a Wald test, and the ellipsoid that defines the boundary of the set computed numerically. A wide range of linear and nonlinear time series models—encompassing long memory, state space, and mixture specifications—is used to demonstrate the procedure, based on artificially generated data. An empirical example in which distributional forecasts of both financial returns and its stochastic volatility are produced is then used to illustrate the practical importance of accommodating sampling variation in the manner proposed.

KW - Animated graphics

KW - Confidence regions

KW - Heterogeneous autoregressive model

KW - Probabilistic forecasts

KW - Realized volatility

KW - Time series models

UR - http://www.scopus.com/inward/record.url?scp=85052157686&partnerID=8YFLogxK

U2 - 10.1080/10618600.2018.1476252

DO - 10.1080/10618600.2018.1476252

M3 - Article

VL - 28

SP - 92

EP - 104

JO - Journal of Computational and Graphical Statistics

JF - Journal of Computational and Graphical Statistics

SN - 1061-8600

IS - 1

ER -