Abstract
By introducing a genetic algorithm learning with a classifier system, we construct an AI model for oil volatility forecasting on the basis of Internal Information and External Information. The model provides decision support for mark-to-market portfolio and risk management by forecasting whether 1-day-ahead volatility is above a given threshold. Moreover, we explore the dynamic influencing mechanism of different types of information through information usage frequency in the learning process. In particular, we find that the jump component of oil realized volatility is efficient only in the bull market, and currency information contributes most rather than oil information in the bear market. Therefore, this article provides an AI method to forecast oil volatility as well as to improve the information structure of forecasting models.
-
Author Keywords
- Oil volatility,
- forecasting,
- genetic algorithm,
- AI,
- classifier system
-
IEEE Keywords
- Forecasting,
- Predictive models,
- Indexes,
- Intelligent systems,
- Genetic algorithms
Introduction
Substantial fluctuations in crude oil prices have always attracted the attention of market traders because oil price volatility is important for portfolio optimization, options and derivatives pricing, value-at-risk modeling, risk management, and he dging[1]. The majority of the existing contributions oil volatility forecasting uses GARCH-style models based on the information on historical oil prices. However, Wangand Liu[2]find that the oil market is not efficient in its weak form. This finding means that historical oil prices can not fully reflect macroeconomic uncertainty and introducing external information into oil volatility forecasting models necessary. Moreover, conventional arch-style methods focus on optimizing the overall fitness of models to a particular data sample set rather than the forecasting accuracy of a specific date, although the latter is more useful for mark-to-market decision support. Its also assumed thattheinfluencingmechanism of discussed information on oil volatility is constant over long periods of time. Ma et al.[3]construct four predictors based on different information sources and find the differences of applicability of these predictors in different market periods and forecasting horizons. This result shows the driving mechanism of oil volatility is dynamic rather than fixed. Other efforts are based on artificial intelligence (AI) models, especially on neural networkmodels[4]. These methods usually perform excellently in the training set while the results lack interpretability and are difficult to provide theoretical implications for financial risk management in the industry. Therefore, we build the AI model by employing a genetic algorithm (GA) learning with a classifier system. We introduce external information into the model to improve forecasting accuracy. With the help of evolving classifier rule set and GA learning based on recent forecasting performance, the model gives reliable predictions of 1-day-ahead oil volatility and can support decision making on mark-to-market scenarios. Further, we explore the dynamic influencing mechanism through information usage frequency and find a way to improve the information structure of oil volatility forecasting.
Our contributions include two parts. First, unlike conventional methods that pursue minimizing the average fitting error of the observation period, we introduce the AI method for oil volatility forecasting based on internal and external information. Moreover, this model focuses on real-time analysis and is adaptive to the changing market mechanism, thus it opens the “black box” of information usage mechanism of oil volatility forecasting and the results give implication on adjusting the information structure of forecasting models.
Conclusion
The model in this study builds its forecasting ability on the basis of self-learning and it’s adaptive to the changing market dynamics. The AI model aims at optimizing the forecasting accuracy of specific dates on the basis of dynamic and real-time analysis. Thus, the forecasting results can directly support the decision-making of risk management. With the help of information usage frequency, the model characterizes different influencing mechanisms of oil volatility under various market situations and provides implications for improving information structure of volatility forecasting models
Acknowledgment
This work is supported by the National Natural Science Foundation of China(U1811462 and 71671191).
About KSRA
The Kavian Scientific Research Association (KSRA) is a non-profit research organization to provide research / educational services in December 2013. The members of the community had formed a virtual group on the Viber social network. The core of the Kavian Scientific Association was formed with these members as founders. These individuals, led by Professor Siavosh Kaviani, decided to launch a scientific / research association with an emphasis on education.
KSRA research association, as a non-profit research firm, is committed to providing research services in the field of knowledge. The main beneficiaries of this association are public or private knowledge-based companies, students, researchers, researchers, professors, universities, and industrial and semi-industrial centers around the world.
Our main services Based on Education for all Spectrum people in the world. We want to make an integration between researches and educations. We believe education is the main right of Human beings. So our services should be concentrated on inclusive education.
The KSRA team partners with local under-served communities around the world to improve the access to and quality of knowledge based on education, amplify and augment learning programs where they exist, and create new opportunities for e-learning where traditional education systems are lacking or non-existent.
FULL Paper PDF file:
An AI Model for Oil Volatility ForecastingBibliography
author
Year
2020
Title
An AI Model for Oil Volatility Forecasting
Publish in
in IEEE Intelligent Systems, vol. 35, no. 3, pp. 62-70, 1 May-June 2020
Doi
10.1109/MIS.2020.2972544.
PDF reference and original file: Click here
Somayeh Nosrati was born in 1982 in Tehran. She holds a Master's degree in artificial intelligence from Khatam University of Tehran.
Professor Siavosh Kaviani was born in 1961 in Tehran. He had a professorship. He holds a Ph.D. in Software Engineering from the QL University of Software Development Methodology and an honorary Ph.D. from the University of Chelsea.
Nasim Gazerani was born in 1983 in Arak. She holds a Master's degree in Software Engineering from UM University of Malaysia.
