Data analytics enhanced component volatility model

Yuan Yao, Jia Zhai, Yi Cao, Xuemei Ding, Junxiu Liu, Yuling Luo

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Volatility modelling and forecasting have attracted many attentions in both finance and computation areas. Recent advances in machine learning allow us to construct complex models on volatility forecasting. However, the machine learning algorithms have been used merely as additional tools to the existing econometrics models. The hybrid models that specifically capture the characteristics of the volatility data have not been developed yet. We propose a new hybrid model, which is constructed by a low-pass filter, the autoregressive neural network and an autoregressive model. The volatility data is decomposed by the low-pass filter into long and short term components, which are then modelled by the autoregressive neural network and an autoregressive model respectively. The total forecasting result is aggregated by the outputs of two models. The experimental evaluations using one-hour and one-day realized volatility across four major foreign exchanges showed that the proposed model significantly outperforms the component GARCH, EGARCH and neural network only models in all forecasting horizons.
LanguageEnglish
Pages232-241
Number of pages10
JournalExpert Systems with Applications
Volume84
Early online date10 May 2017
DOIs
Publication statusPublished - 30 Oct 2017

Fingerprint

Neural networks
Volatility models
Hybrid model
Volatility forecasting
Filter
Autoregressive model
Machine learning
Learning algorithm
Finance
Network model
Realized volatility
Econometric models
Foreign exchange
Evaluation
Generalized autoregressive conditional heteroscedasticity
Volatility modelling

Keywords

  • Autoregressive neural network
  • Hybrid model
  • Two-component
  • Volatility model

Cite this

Yao, Yuan ; Zhai, Jia ; Cao, Yi ; Ding, Xuemei ; Liu, Junxiu ; Luo, Yuling. / Data analytics enhanced component volatility model. In: Expert Systems with Applications. 2017 ; Vol. 84. pp. 232-241.
@article{cc14f31fa96549e1aaa89643e5c8d98f,
title = "Data analytics enhanced component volatility model",
abstract = "Volatility modelling and forecasting have attracted many attentions in both finance and computation areas. Recent advances in machine learning allow us to construct complex models on volatility forecasting. However, the machine learning algorithms have been used merely as additional tools to the existing econometrics models. The hybrid models that specifically capture the characteristics of the volatility data have not been developed yet. We propose a new hybrid model, which is constructed by a low-pass filter, the autoregressive neural network and an autoregressive model. The volatility data is decomposed by the low-pass filter into long and short term components, which are then modelled by the autoregressive neural network and an autoregressive model respectively. The total forecasting result is aggregated by the outputs of two models. The experimental evaluations using one-hour and one-day realized volatility across four major foreign exchanges showed that the proposed model significantly outperforms the component GARCH, EGARCH and neural network only models in all forecasting horizons.",
keywords = "Autoregressive neural network, Hybrid model, Two-component, Volatility model",
author = "Yuan Yao and Jia Zhai and Yi Cao and Xuemei Ding and Junxiu Liu and Yuling Luo",
note = "Compliant at Salford University - evidence uploaded to other files",
year = "2017",
month = "10",
day = "30",
doi = "10.1016/j.eswa.2017.05.025",
language = "English",
volume = "84",
pages = "232--241",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "Elsevier",

}

Data analytics enhanced component volatility model. / Yao, Yuan; Zhai, Jia; Cao, Yi; Ding, Xuemei; Liu, Junxiu; Luo, Yuling.

In: Expert Systems with Applications, Vol. 84, 30.10.2017, p. 232-241.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Data analytics enhanced component volatility model

AU - Yao, Yuan

AU - Zhai, Jia

AU - Cao, Yi

AU - Ding, Xuemei

AU - Liu, Junxiu

AU - Luo, Yuling

N1 - Compliant at Salford University - evidence uploaded to other files

PY - 2017/10/30

Y1 - 2017/10/30

N2 - Volatility modelling and forecasting have attracted many attentions in both finance and computation areas. Recent advances in machine learning allow us to construct complex models on volatility forecasting. However, the machine learning algorithms have been used merely as additional tools to the existing econometrics models. The hybrid models that specifically capture the characteristics of the volatility data have not been developed yet. We propose a new hybrid model, which is constructed by a low-pass filter, the autoregressive neural network and an autoregressive model. The volatility data is decomposed by the low-pass filter into long and short term components, which are then modelled by the autoregressive neural network and an autoregressive model respectively. The total forecasting result is aggregated by the outputs of two models. The experimental evaluations using one-hour and one-day realized volatility across four major foreign exchanges showed that the proposed model significantly outperforms the component GARCH, EGARCH and neural network only models in all forecasting horizons.

AB - Volatility modelling and forecasting have attracted many attentions in both finance and computation areas. Recent advances in machine learning allow us to construct complex models on volatility forecasting. However, the machine learning algorithms have been used merely as additional tools to the existing econometrics models. The hybrid models that specifically capture the characteristics of the volatility data have not been developed yet. We propose a new hybrid model, which is constructed by a low-pass filter, the autoregressive neural network and an autoregressive model. The volatility data is decomposed by the low-pass filter into long and short term components, which are then modelled by the autoregressive neural network and an autoregressive model respectively. The total forecasting result is aggregated by the outputs of two models. The experimental evaluations using one-hour and one-day realized volatility across four major foreign exchanges showed that the proposed model significantly outperforms the component GARCH, EGARCH and neural network only models in all forecasting horizons.

KW - Autoregressive neural network

KW - Hybrid model

KW - Two-component

KW - Volatility model

U2 - 10.1016/j.eswa.2017.05.025

DO - 10.1016/j.eswa.2017.05.025

M3 - Article

VL - 84

SP - 232

EP - 241

JO - Expert Systems with Applications

T2 - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

ER -