Accelerating score-driven time series models

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

We propose a new class of score-driven time series models that allows for a more flexible weighting of score innovations for the filtering of time varying parameters. The parameter for the score innovation is made time-varying by means of an updating equation that accounts for the autocorrelations of past innovations. We provide the theoretical foundations for this acceleration method by showing optimality in terms of reducing Kullback–Leibler divergence. The empirical relevance of this accelerated score-driven updating method is illustrated in two empirical studies. First, we include acceleration in the generalized autoregressive conditional heteroskedasticity model. We adopt the new model to extract volatility from exchange rates and to analyze daily density forecasts of volatilities from all individual stock return series in the Standard & Poor's 500 index. Second, we consider a score-driven acceleration for the time-varying mean and use this new model in a forecasting study for US inflation.

Original languageEnglish
JournalJournal of Econometrics
DOIs
Publication statusAccepted/In press - 2019

Fingerprint

Time series models
Innovation
Time-varying
Autocorrelation
Weighting
Autoregressive conditional heteroskedasticity
Optimality
Inflation
Kullback-Leibler divergence
Time-varying parameters
Stock returns
Exchange rates
Empirical study
Density forecasts

Keywords

  • GARCH models
  • Kullback–Leibler divergence
  • S&P 500 stocks
  • Score-driven models
  • Time-varying parameters
  • US inflation

Cite this

@article{d51116ec1160473ebbe055187568d7c5,
title = "Accelerating score-driven time series models",
abstract = "We propose a new class of score-driven time series models that allows for a more flexible weighting of score innovations for the filtering of time varying parameters. The parameter for the score innovation is made time-varying by means of an updating equation that accounts for the autocorrelations of past innovations. We provide the theoretical foundations for this acceleration method by showing optimality in terms of reducing Kullback–Leibler divergence. The empirical relevance of this accelerated score-driven updating method is illustrated in two empirical studies. First, we include acceleration in the generalized autoregressive conditional heteroskedasticity model. We adopt the new model to extract volatility from exchange rates and to analyze daily density forecasts of volatilities from all individual stock return series in the Standard & Poor's 500 index. Second, we consider a score-driven acceleration for the time-varying mean and use this new model in a forecasting study for US inflation.",
keywords = "GARCH models, Kullback–Leibler divergence, S&P 500 stocks, Score-driven models, Time-varying parameters, US inflation",
author = "F. Blasques and P. Gorgi and Koopman, {S. J.}",
year = "2019",
doi = "10.1016/j.jeconom.2019.03.005",
language = "English",
journal = "Journal of Econometrics",
issn = "0304-4076",
publisher = "Elsevier BV",

}

Accelerating score-driven time series models. / Blasques, F.; Gorgi, P.; Koopman, S. J.

In: Journal of Econometrics, 2019.

Research output: Contribution to JournalArticleAcademicpeer-review

TY - JOUR

T1 - Accelerating score-driven time series models

AU - Blasques, F.

AU - Gorgi, P.

AU - Koopman, S. J.

PY - 2019

Y1 - 2019

N2 - We propose a new class of score-driven time series models that allows for a more flexible weighting of score innovations for the filtering of time varying parameters. The parameter for the score innovation is made time-varying by means of an updating equation that accounts for the autocorrelations of past innovations. We provide the theoretical foundations for this acceleration method by showing optimality in terms of reducing Kullback–Leibler divergence. The empirical relevance of this accelerated score-driven updating method is illustrated in two empirical studies. First, we include acceleration in the generalized autoregressive conditional heteroskedasticity model. We adopt the new model to extract volatility from exchange rates and to analyze daily density forecasts of volatilities from all individual stock return series in the Standard & Poor's 500 index. Second, we consider a score-driven acceleration for the time-varying mean and use this new model in a forecasting study for US inflation.

AB - We propose a new class of score-driven time series models that allows for a more flexible weighting of score innovations for the filtering of time varying parameters. The parameter for the score innovation is made time-varying by means of an updating equation that accounts for the autocorrelations of past innovations. We provide the theoretical foundations for this acceleration method by showing optimality in terms of reducing Kullback–Leibler divergence. The empirical relevance of this accelerated score-driven updating method is illustrated in two empirical studies. First, we include acceleration in the generalized autoregressive conditional heteroskedasticity model. We adopt the new model to extract volatility from exchange rates and to analyze daily density forecasts of volatilities from all individual stock return series in the Standard & Poor's 500 index. Second, we consider a score-driven acceleration for the time-varying mean and use this new model in a forecasting study for US inflation.

KW - GARCH models

KW - Kullback–Leibler divergence

KW - S&P 500 stocks

KW - Score-driven models

KW - Time-varying parameters

KW - US inflation

UR - http://www.scopus.com/inward/record.url?scp=85064814141&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85064814141&partnerID=8YFLogxK

U2 - 10.1016/j.jeconom.2019.03.005

DO - 10.1016/j.jeconom.2019.03.005

M3 - Article

JO - Journal of Econometrics

JF - Journal of Econometrics

SN - 0304-4076

ER -