Discriminating modelling approaches for Point in Time Economic Scenario Generation

We introduce the notion of Point in Time Economic Scenario Generation (PiT ESG) with a clear mathematical problem formulation to unify and compare economic scenario generation approaches conditional on forward looking market data. Such PiT ESGs should provide quicker and more flexible reactions to sudden economic changes than traditional ESGs calibrated solely to long periods of historical data. We specifically take as economic variable the S&P500 Index with the VIX Index as forward looking market data to compare the nonparametric filtered historical simulation, GARCH model with joint likelihood estimation (parametric), Restricted Boltzmann Machine and the conditional Variational Autoencoder (Generative Networks) for their suitability as PiT ESG. Our evaluation consists of statistical tests for model fit and benchmarking the out of sample forecasting quality with a strategy backtest using model output as stop loss criterion. We find that both Generative Networks outperform the nonparametric and classic parametric model in our tests, but that the CVAE seems to be particularly well suited for our purposes: yielding more robust performance and being computationally lighter.

[1]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[2]  Misha van Beek,et al.  Consistent Calibration of Economic Scenario Generators: The Case for Conditional Simulation , 2020, 2004.09042.

[3]  Geoffrey E. Hinton.  Training Products of Experts by Minimizing Contrastive Divergence , 2002, Neural Computation.

[4]  Robert E. Whaley,et al.  The Investor Fear Gauge , 2000 .

[5]  Thierry Roncalli,et al.  Improving the Robustness of Trading Strategy Backtesting with Boltzmann Machines and Generative Adversarial Networks , 2020, SSRN Electronic Journal.

[6]  B. W. Golub,et al.  Market-Driven Scenarios: An Approach for Plausible Scenario Construction , 2018 .

[7]  Peter Carr,et al.  A Tale of Two Indices , 2004 .

[8]  B. Adrangi,et al.  Dynamic Responses of Major Equity Markets to the US Fear Index , 2019, Journal of Risk and Financial Management.

[9]  D. Hendricks,et al.  Evaluation of Value-at-Risk Models Using Historical Data , 1996 .

[10]  A. Lo,et al.  When Do Stop-Loss Rules Stop Losses? , 2014 .

[11]  J. MacKinnon,et al.  Maximum Likelihood Estimation of Singular Equation Systems with Autoregressive Disturbances , 1979 .

[12]  Alan G. White,et al.  INCORPORATING VOLATILITY UPDATING INTO THE HISTORICAL SIMULATION METHOD FOR VALUE AT RISK , 1998 .

[13]  Thomas W. Miller,et al.  The Forecast Quality of Cboe Implied Volatility Indexes , 2003 .

[14]  Jin E. Zhang,et al.  GARCH Option Pricing Models, the CBOE VIX, and Variance Risk Premium , 2013 .

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

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

[17]  M. V. D. Schans,et al.  Time-Dependent Black–Litterman , 2017 .

[18]  Marco Lo Duca,et al.  CISS - A Composite Indicator of Systemic Stress in the Financial System , 2012, SSRN Electronic Journal.

[19]  Philip C. Treleaven,et al.  Generative adversarial networks for financial trading strategies fine-tuning and combination , 2019, Quantitative Finance.

[21]  Financial Time Series: Stylized Facts for the Mexican Stock Exchange Index Compared to Developed Markets , 2014, 1412.3126.

[22]  H. Markovitz.  The Likelihood of Various Stock Market Return Distributions, Part 2: Empirical Results , 1997 .

[23]  R. Cont.  Empirical properties of asset returns: stylized facts and statistical issues , 2001 .

[24]  Gianluca Cerminara,et al.  Anomaly Detection with Conditional Variational Autoencoders , 2019, 2019 18th IEEE International Conference On Machine Learning And Applications (ICMLA).

[25]  Giovanni Barone-Adesi,et al.  VaR without correlations for portfolios of derivative securities , 1999 .

[26]  E. Platen,et al.  The marginal distributions of returns and volatility , 1997 .

[27]  Tanya Styblo Beder,et al.  VAR: Seductive but Dangerous , 1995 .

[28]  Lars Stentoft,et al.  Affine multivariate GARCH models , 2020 .

[29]  M. V. D. Schans,et al.  Views, Factor Models and Optimal Asset Allocation , 2015 .

[30]  Geoffrey E. Hinton,et al.  Two Distributed-State Models For Generating High-Dimensional Time Series , 2011, J. Mach. Learn. Res..

[31]  Christian Igel,et al.  An Introduction to Restricted Boltzmann Machines , 2012, CIARP.

[32]  J. Duan.  THE GARCH OPTION PRICING MODEL , 1995 .

[33]  Evgueni A. Haroutunian,et al.  Information Theory and Statistics , 2011, International Encyclopedia of Statistical Science.

[34]  Alexander J. McNeil,et al.  Integrated models of capital adequacy – Why banks are undercapitalised , 2010 .

[35]  Kaplan,et al.  ‘Combining Probability Distributions from Experts in Risk Analysis’ , 2000, Risk analysis : an official publication of the Society for Risk Analysis.

[36]  Alexei Kondratyev,et al.  Learning Curve Dynamics with Artificial Neural Networks , 2018 .

[37]  Alexander J. McNeil,et al.  Quantitative Risk Management: Concepts, Techniques and Tools Revised edition , 2015 .

[38]  Paul Smolensky,et al.  Information processing in dynamical systems: foundations of harmony theory , 1986 .

[39]  Bernhard Schölkopf,et al.  From Variational to Deterministic Autoencoders , 2019, ICLR.

[40]  David Hinkley,et al.  Bootstrap Methods: Another Look at the Jackknife , 2008 .

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

[42]  S. Heston.  A Closed-Form Solution for Options with Stochastic Volatility with Applications to Bond and Currency Options , 1993 .

[43]  Geoffrey E. Hinton,et al.  Restricted Boltzmann machines for collaborative filtering , 2007, ICML '07.

[44]  A. Meucci.  Risk and asset allocation , 2005 .

[45]  J. Kanniainen,et al.  Estimating and Using GARCH Models with VIX Data for Option Valuation , 2014 .

[46]  P. Giot.  Relationships Between Implied Volatility Indexes and Stock Index Returns , 2005 .

[47]  L. Böttcher,et al.  Learning the Ising Model with Generative Neural Networks , 2020, Physical Review Research.

[48]  Max Welling,et al.  Auto-Encoding Variational Bayes , 2013, ICLR.

[49]  Robert B. Litterman,et al.  Global Portfolio Optimization , 1992 .

[50]  F. Black,et al.  The Pricing of Options and Corporate Liabilities , 1973, Journal of Political Economy.

[51]  Anders Vilhelmsson,et al.  Nonparametric Forward-Looking Value-at-Risk , 2014 .

[52]  E. Varnell.  Economic Scenario Generators and Solvency II , 2011, British Actuarial Journal.

[53]  Thomas J. Latta.  Stress Testing in a Value at Risk Framework , 1999 .

[54]  Michael T. Owyang,et al.  Disentangling diverse measures: a survey of financial stress indexes , 2012 .

[55]  Edward Qian,et al.  Conditional Distribution in Portfolio Theory , 2001 .

[56]  Honglak Lee,et al.  Learning Structured Output Representation using Deep Conditional Generative Models , 2015, NIPS.