A hybrid volatility forecasting framework integrating GARCH, artificial neural network, technical analysis and principal components analysis

Abstract Measurement, prediction, and modeling of currency price volatility constitutes an important area of research at both the national and corporate level. Countries attempt to understand currency volatility to set national economic policies and firms to best manage exchange rate risk and leverage assets. A relatively new technological invention that the corporate treasurer has to turn to as part of the overall financial strategy is cryptocurrency. One estimate values the total market capitalization of cryptocurrencies at $557 billion USD at the beginning of 2018. While the overall size of the market for cryptocurrency is significant, our understanding of the behavior of this instrument is only beginning. In this article, we propose a hybrid Artificial Neural Network-Generalized AutoRegressive Conditional Heteroskedasticity (ANN-GARCH) model with preprocessing to forecast the price volatility of bitcoin, the most traded and largest by market capitalization of the cryptocurrencies.

[1]  A. H. Dyhrberg.  Bitcoin, gold and the dollar – A GARCH volatility analysis , 2016 .

[2]  Min Qi,et al.  Nonlinear Predictability of Stock Returns Using Financial and Economic Variables , 1999 .

[3]  Vivek K. Pandey,et al.  The Value of Bitcoin in Enhancing the Efficiency of an Investor's Portfolio , 2014 .

[4]  R. Engle.  GARCH 101: The Use of ARCH/GARCH Models in Applied Econometrics , 2001 .

[5]  B. Mccallum.  The Bitcoin Revolution , 2015 .

[6]  G. Box,et al.  Distribution of Residual Autocorrelations in Autoregressive-Integrated Moving Average Time Series Models , 1970 .

[7]  H. Qu,et al.  Adaptive Heterogeneous Autoregressive Models of Realized Volatility Based on a Genetic Algorithm , 2014 .

[8]  Daniel B. Nelson.  CONDITIONAL HETEROSKEDASTICITY IN ASSET RETURNS: A NEW APPROACH , 1991 .

[9]  Xin Li,et al.  The Technology and Economic Determinants of Cryptocurrency Exchange Rates: The Case of Bitcoin , 2014, Decis. Support Syst..

[10]  A. Szafarz,et al.  Virtual currency, tangible return: Portfolio diversification with bitcoin , 2015 .

[11]  G. Maddala,et al.  Economic factors and the stock market: a new perspective , 1999 .

[12]  Sheng-Tzong Cheng,et al.  Artificial neural network model of the hybrid EGARCH volatility of the Taiwan stock index option prices , 2008 .

[13]  Ladislav Kristoufek,et al.  What Are the Main Drivers of the Bitcoin Price? Evidence from Wavelet Coherence Analysis , 2014, PloS one.

[14]  Asger Lunde,et al.  Choosing the Best Volatility Models: The Model Confidence Set Approach , 2003 .

[15]  A. Cheung,et al.  Crypto-currency bubbles: an application of the Phillips–Shi–Yu (2013) methodology on Mt. Gox bitcoin prices , 2015 .

[16]  H. White,et al.  Economic prediction using neural networks: the case of IBM daily stock returns , 1988, IEEE 1988 International Conference on Neural Networks.

[17]  Arash Bahrammirzaee,et al.  A comparative survey of artificial intelligence applications in finance: artificial neural networks, expert system and hybrid intelligent systems , 2010, Neural Computing and Applications.

[18]  Xunfa Lu,et al.  Volatility Forecast Based on the Hybrid Artificial Neural Network and GARCH-type Models , 2016 .

[19]  Taisei Kaizoji,et al.  Volatility Analysis of Bitcoin Price Time Series , 2017 .

[20]  James D. Hamilton,et al.  Autoregressive conditional heteroskedasticity and changes in regime , 1994 .

[21]  Werner Kristjanpoller,et al.  Volatility forecast using hybrid Neural Network models , 2014, Expert Syst. Appl..

[22]  R. Gencay.  Non-linear prediction of security returns with moving average rules , 1996 .

[23]  T. Bollerslev,et al.  Generalized autoregressive conditional heteroskedasticity , 1986 .

[24]  R. Donaldson,et al.  An artificial neural network-GARCH model for international stock return volatility , 1997 .

[25]  Stephen L Taylor,et al.  Modelling Financial Time Series , 1987 .

[26]  P. Whittle.  On principal components and least square methods of factor analysis , 2012 .

[27]  C. Granger,et al.  A long memory property of stock market returns and a new model , 1993 .

[28]  Md. Ashraful Islam Khan.  Financial Volatility Forecasting by Nonlinear Support Vector Machine Heterogeneous Autoregressive Model: Evidence from Nikkei 225 Stock Index , 2011 .

[29]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[30]  G. Box,et al.  On a measure of lack of fit in time series models , 1978 .

[31]  Melike Bildirici,et al.  Forecasting oil prices: Smooth transition and neural network augmented GARCH family models , 2013 .

[32]  Melike Bildirici,et al.  Improving forecasts of GARCH family models with the artificial neural networks: An application to the daily returns in Istanbul Stock Exchange , 2009, Expert Syst. Appl..

[33]  Jui-Chung Hung,et al.  Adaptive Fuzzy-GARCH model applied to forecasting the volatility of stock markets using particle swarm optimization , 2011, Inf. Sci..

[34]  Werner Kristjanpoller,et al.  Gold price volatility: A forecasting approach using the Artificial Neural Network-GARCH model , 2015, Expert Syst. Appl..

[35]  Werner Kristjanpoller,et al.  Forecasting volatility of oil price using an artificial neural network-GARCH model , 2016, Expert Syst. Appl..

[36]  Peter Reinhard Hansen,et al.  The Model Confidence Set , 2010 .

[37]  Steve Y. Yang,et al.  Bitcoin Market Return and Volatility Forecasting Using Transaction Network Flow Properties , 2015 .

[38]  Timo Teräsvirta,et al.  POWER OF THE NEURAL NETWORK LINEARITY TEST , 1993 .

[39]  Jon Carrick.  Bitcoin as a Complement to Emerging Market Currencies , 2016 .

[40]  M. Medeiros,et al.  Modeling and predicting the CBOE market volatility index , 2014 .

[41]  Yiu Kuen Tse,et al.  The conditional heteroscedasticity of the yen-dollar exchange rate , 1998 .

[42]  Marshall W. van Alstyne,et al.  Why Bitcoin has value , 2014, CACM.

[43]  David Enke,et al.  Volatility Forecasting Using a Hybrid GJR-GARCH Neural Network Model , 2014, Complex Adaptive Systems.

[44]  Marwan Izzeldin,et al.  On forecasting daily stock volatility: The role of intraday information and market conditions , 2009 .

[45]  Mohammad Hossein Fazel Zarandi,et al.  A hybrid modeling approach for forecasting the volatility of S&P 500 index return , 2012, Expert Syst. Appl..

[46]  R WernerKristjanpoller,et al.  Volatility of main metals forecasted by a hybrid ANN-GARCH model with regressors , 2017 .

[47]  Fernando A. C. Gomide,et al.  Evolving Hybrid Neural Fuzzy Network for System Modeling and Time Series Forecasting , 2013, 2013 12th International Conference on Machine Learning and Applications.

[48]  E. Dockner,et al.  Forecasting Time-dependent Conditional Densities: A Semi-non- parametric Neural Network Approach , 2000 .

[49]  James Davidson,et al.  Moment and Memory Properties of Linear Conditional Heteroscedasticity Models, and a New Model , 2004 .

[50]  Tugrul U. Daim,et al.  Using artificial neural network models in stock market index prediction , 2011, Expert Syst. Appl..

[51]  Timothy (Jun) Lu,et al.  Capital Flight and Bitcoin Regulation , 2016 .

[52]  T. Moore,et al.  Bitcoin: Economics, Technology, and Governance , 2014 .

[53]  David W. Wright,et al.  Using Artificial Neural Networks to Pick Stocks , 1993 .

[54]  Geoffrey Lightfoot,et al.  Price Fluctuations and the Use of Bitcoin: An Empirical Inquiry , 2015, Int. J. Electron. Commer..

[55]  Tae Hyup Roh.  Forecasting the volatility of stock price index , 2007, Expert Syst. Appl..