Realized recurrent conditional heteroskedasticity model for volatility modelling

We propose a new approach to volatility modelling by combining deep learning (LSTM) and realized volatility measures. This LSTM-enhanced realized GARCH framework incorporates and distills modeling advances from financial econometrics, high frequency trading data and deep learning. Bayesian inference via the Sequential Monte Carlo method is employed for statistical inference and forecasting. The new framework can jointly model the returns and realized volatility measures, has an excellent in-sample fit and superior predictive performance compared to several benchmark models, while being able to adapt well to the stylized facts in volatility. The performance of the new framework is tested using a wide range of metrics, from marginal likelihood, volatility forecasting, to tail risk forecasting and option pricing. We report on a comprehensive empirical study using 31 widely traded stock indices over a time period that includes COVID-19 pandemic.

[1]  R. Kohn,et al.  Flexible and Robust Particle Tempering for State Space Models , 2022, Econometrics and Statistics.

[2]  Eghbal Rahimikia,et al.  Realised Volatility Forecasting: Machine Learning via Financial Word Embedding , 2021, SSRN Electronic Journal.

[3]  R. Kohn,et al.  Recurrent Conditional Heteroskedasticity † , 2020, Journal of Applied Econometrics.

[4]  Andrea Bucci,et al.  Realized Volatility Forecasting with Neural Networks , 2020, Journal of Financial Econometrics.

[5]  Haibin Xie,et al.  Realized GARCH models: Simpler is better , 2020 .

[6]  Yang Liu,et al.  Novel volatility forecasting using deep learning-Long Short Term Memory Recurrent Neural Networks , 2019, Expert Syst. Appl..

[7]  Erik Cambria,et al.  Sentiment-aware volatility forecasting , 2019, Knowl. Based Syst..

[8]  Dan Li,et al.  Efficient Bayesian estimation for GARCH-type models via Sequential Monte Carlo , 2019, Econometrics and Statistics.

[9]  James W. Taylor.  Forecasting Value at Risk and Expected Shortfall Using a Semiparametric Approach Based on the Asymmetric Laplace Distribution , 2019 .

[10]  Ha Young Kim,et al.  Forecasting the volatility of stock price index: A hybrid model integrating LSTM with multiple GARCH-type models , 2018, Expert Syst. Appl..

[11]  Ye Li,et al.  Modeling returns volatility: Realized GARCH incorporating realized risk measure , 2018, Physica A: Statistical Mechanics and its Applications.

[12]  Adam Atkins,et al.  Financial news predicts stock market volatility better than close price , 2018, The Journal of Finance and Data Science.

[13]  Richard Gerlach,et al.  Forecasting risk via realized GARCH, incorporating the realized range , 2016 .

[14]  Peter Reinhard Hansen,et al.  Exponential GARCH Modeling With Realized Measures of Volatility , 2015 .

[15]  Johanna F. Ziegel,et al.  Higher order elicitability and Osband’s principle , 2015, 1503.08123.

[16]  Arnaud Doucet,et al.  An adaptive sequential Monte Carlo method for approximate Bayesian computation , 2011, Statistics and Computing.

[17]  P. Hansen,et al.  Realized GARCH: A Joint Model of Returns and Realized Measures of Volatility , 2010 .

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

[19]  Neil Shephard,et al.  Realising the future: forecasting with high frequency based volatility (HEAVY) models , 2010 .

[20]  Fulvio Corsi,et al.  A Simple Approximate Long-Memory Model of Realized Volatility , 2008 .

[21]  Tim Bollerslev,et al.  Glossary to ARCH (GARCH) , 2008 .

[22]  Neil Shephard,et al.  Designing Realised Kernels to Measure the Ex-Post Variation of Equity Prices in the Presence of Noise , 2008 .

[23]  B. Christensen,et al.  The Effect of Long Memory in Volatility on Stock Market Fluctuations , 2007, The Review of Economics and Statistics.

[24]  Piotr Cofta,et al.  The Model of Confidence , 2007 .

[25]  Tony O’Hagan.  Bayes factors , 2006 .

[26]  Tae Hyup Roh,et al.  Forecasting the volatility of stock price index , 2006, Expert Syst. Appl..

[27]  P. Hansen,et al.  A Forecast Comparison of Volatility Models: Does Anything Beat a Garch(1,1)? , 2004 .

[28]  R. Engle,et al.  A Multiple Indicators Model for Volatility Using Intra-Daily Data , 2003 .

[29]  Tim Bollerslev,et al.  Bridging the gap between the distribution of realized (ECU) volatility and ARCH modelling (of the Euro): the GARCH‐NIG model , 2002 .

[30]  N. Chopin.  A sequential particle filter method for static models , 2002 .

[31]  R. Engle.  New Frontiers for Arch Models , 2002 .

[32]  Francis X. Diebold,et al.  Modeling and Forecasting Realized Volatility , 2001 .

[33]  T. Bollerslev,et al.  ANSWERING THE SKEPTICS: YES, STANDARD VOLATILITY MODELS DO PROVIDE ACCURATE FORECASTS* , 1998 .

[34]  Bradley P. Carlin,et al.  Markov Chain Monte Carlo in Practice: A Roundtable Discussion , 1998 .

[35]  Radford M. Neal.  Annealed importance sampling , 1998, Stat. Comput..

[36]  S. Hochreiter,et al.  Long Short-Term Memory , 1997, Neural Computation.

[37]  L. Glosten,et al.  On the Relation between the Expected Value and the Volatility of the Nominal Excess Return on Stocks , 1993 .

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

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

[40]  R. Engle,et al.  Valuation of Variance Forecast with Simulated Option Markets , 1990 .

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

[42]  R. Engle.  Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation , 1982 .

[43]  R. Koenker,et al.  Regression Quantiles , 2007 .

[44]  N. Shephard,et al.  Power and bipower variation with stochastic volatility and jumps , 2003 .

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

[46]  B. Mandelbrot.  The Variation of Some Other Speculative Prices , 1967 .

[47]  L. M. M.-T..  Theory of Probability , 1929, Nature.