Dynamic functional time-series forecasts of foreign exchange implied volatility surfaces

Han Lin Shang, Fearghal Kearney

Research output: Contribution to journalArticlepeer-review

6 Citations (Scopus)
63 Downloads (Pure)

Abstract

This paper presents static and dynamic versions of univariate, multivariate, and multilevel functional time-series methods to forecast implied volatility surfaces in foreign exchange markets. We find that dynamic functional principal component analysis generally improves out-of-sample forecast accuracy. Specifically, the dynamic univariate functional time-series method shows the greatest improvement. Our models lead to multiple instances of statistically significant improvements in forecast accuracy for daily EUR–USD, EUR–GBP, and EUR–JPY implied volatility surfaces across various maturities, when benchmarked against established methods. A stylised trading strategy is also employed to demonstrate the potential economic benefits of our proposed approach.
Original languageEnglish
Pages (from-to)1025-1049
Number of pages25
JournalInternational Journal of Forecasting
Volume38
Issue number3
Early online date10 Jun 2022
DOIs
Publication statusPublished - 01 Jul 2022

Bibliographical note

Publisher Copyright:
© 2021 International Institute of Forecasters

Keywords

  • Augmented common factor method
  • Functional principal component analysis
  • Long-run covariance
  • Stochastic processes
  • Univariate time-series forecasting

ASJC Scopus subject areas

  • Business and International Management

Fingerprint

Dive into the research topics of 'Dynamic functional time-series forecasts of foreign exchange implied volatility surfaces'. Together they form a unique fingerprint.

Cite this