Applications of machine learning for volatility estimation and quantitative strategies Artur Sepp Quantica Capital AG Swissquote Conference 2018 on Machine Learning in Finance 9 November 2018
Machine Learning for Quant Strategies Theoretical foundations of Machine/Statistical Learning: Approximation vs Estimation error Simplicity vs Complexity Why Alternative Risk Premia products failed Example of supervised learning for selecting volatility models Risk-profile of systematic investment strategies 2
Data Overfitting: many solutions to fit data points locally with different global behaviour 3
Example of perfect in-sample fit for an asset 10 price path 5-degree polynomial trend-line with near perfect explanatory power R^2=98% 8 6 Asset price 5-degree Polynomial Fit 4 2 0 4
Example of out-sample forecast for short voletn How to prevent ML algorithms from falling into this trap? 14 12 10 8 Out-of-sample forecast for Feb 2018 Forecast Return=+45% Short Volatility Exchange Traded Note (XIV) 5-degree Polynomial Forecast 6 4 2 0 05-Feb-16 Realized Return=-95% 05-Feb-18 5
Credit derivatives crisis in October 2008 Quant models for credit derivatives relied on multi-parameter models with linear fits: one parameter for market price of each instrument The models failed to calibrate and work in distressed markets during the Financial Crisis S&P 500 Index Daily Movers Gainers% Losers % WMT 0.30% AIG -23.10% MCD 0.23% MER -19.56% KO 0.11% LEH -19.29% GNW -18.64% ACAS -18.17% GE -17.97% DDR -17.10% HIG -16.86% 6
Alternative risk premia (ARP) crisis in October 2018 ARP is marketed by major banks as market-neutral using overstated back-tests ARP products proliferated from 2015 with estimated AuM $500 bln at mid of 2018 Performance of live ARP products from 2015 has been less spectacular than back-tests HFR Bank Systematic Risk Premia Indices October 2018 YTD Performance Gainers YTD % Losers YTD % Rates Momentum Index 7% Multi-Asset Value Index -61% Credit Multi-Style Index 5% Multi-Asset Volatility Index -34% Rates Value Index 4% Equity Volatility Index -27% Currency Volatility Index 4% Equity Multi-Style Index -26% Credit Carry Index 1% Credit Momentum Index -22% Multi-Asset Multi-Style Index -21% Multi-Asset Index -20% Equity Size Index -19% Multi-Asset Momentum Index -17% Equity Index -16% Equity Quality Index -16% Commodity Volatility Index -14% Equity Carry Index -13% Commodity Multi-Style Index -13% Equity Value Index -13% Commodity Smart Beta Index -11% Equity Momentum Index -11% Equity Smart Beta Index -10% Trend-Following Index -10% Currency Carry Index -8% Credit Index -8% 7
Rich class of decision rules may reduce the approximation error but increases the estimation error Bayesian learning: select the rule with the highest posterior probability but prior probabilities are needed(!) Probably Approximately Correct (PAC) learning: if class D is PAC learnable there exists a finite sample size of for given level of approximation and estimation error Approximation error: the class D may not have good rules Estimation error: we are unable to identify the good rule for prediction from training data Class D of Decision rules Hypothesis h1, h2, Training Data Random observed data Rule Selection Learning 8
Vapnik-Chervonenkis(VS) dimension measures the richness of the class of decision rules VC dimension predicts the bounds of the sample size for PAC learning Example using single-parametric threshold classifier: buy if last return is higher than threshold, sell otherwise: the VC dimension is one 12000 11224 Number of years for PAC learning of the classifier from daily data 10000 8000 10 parameter classifier Single parameter classifier 6000 4000 2000 0 1511 2465 1007 344 531 144 77 322 48 21432 15123 11117 85 13 5% 10% 15% 20% 25% 30% 35% 40% 45% Approximation Error 9
PAC learning using Hierarchy of Decision rules Restricting the richness of the class may improve PAC learning but may increase the approximation error Split the class D of all decision rules into a sequence of classes Di which are PAC learnable VC dimension is a measure of the complexity of rules in class Di Select a rule by minimizing: Approximation Error + Complexity D1=the class with simplest decision rules D2=D1+the class with more complex rules D=D1+D2 + 10
PAC learning for the process of systematic trading includes at least three classes of decision rules Signal Look for predictors with highest scores Portfolio Manage risk allocation and diversification Execution Minimize trading costs and slippage Examples of inconsistent trading processes 1. Signal that works only on one asset: cannot diversify the portfolio 2. Signal that changes too frequently: execution costs can be too high 11
Example of designing strategy for volatility trading: learning hierarchy to reduce the dimensionality Strategy design Volatility Model Parameters Strategy Parameters *Optimal 2-d set Split 2-dimensional problem into two orthogonal 1-dimensional problems Volatility Model Parameters *Optimal 1-d set Strategy Parameters *Optimal 1-d set 12
Model forecast of realized volatility is applied to estimate the volatility risk-premium Relative value volatility trading: Sell/buy options with high/low expected spread and delta-hedge 30% 20% 10% 0% Volatility Risk-Premium=VIX at MonthStart - S&P500 Realized Monthly Volatility -10% -20% -30% Average=4% Minus Standard Deviation=-3% Plus Standard Deviation=10% -40% -50% Mar-86 Mar-88 Mar-90 Mar-92 Mar-94 Mar-96 Mar-98 Mar-00 Mar-02 Mar-04 Mar-06 Mar-08 Mar-10 Mar-12 Mar-14 Mar-16 Mar-18 13
Multiple classes of volatility models are applied for the forecast of realized volatility Sample space estimators GARCH models Bayesian parametric models Close-to-close, Intraday estimators (Parkinson, etc ) Assume random walk for the volatility Garch (1,1), Asymmetric Garch, etc Apply long-term history with mean-reversion Continuous type models with priors for vol forecast Apply intraday high/low price data Hidden Markov Chain Models (HMC) Discrete states of volatility Classification problem in unsupervised machine learning 14
Selection of model with the best forecast power Class of decision rules: all volatility models Implementation: use 40 models from 4 model classes Uniform metric for model selection Implementation: distribution tests for the stability of the forecast Select model with the highest score for the asset or asset class Implementation: Regularly update the tests as new data is available 15
Distribution tests is applied for volatility normalized returns over forecast period 18% Naive: Close-To-Close Volatility Estimator for HY Bonds ETF 18% Advanced: Hidden Markov Chain Volatility Estimator for HY Bonds ETF 16% 14% 12% Empirical Normal (0,1) 16% 14% 12% Empirical Normal (0,1) 10% 10% 8% 8% 6% 6% 4% 4% 2% 2% 0% -6-5 -4-3 -2-1 0 1 2 3 4 5 6 Volatility normalized return 0% -6-5 -4-3 -2-1 0 1 2 3 4 5 6 Volatility normalized return 16
Robust estimator provides tight bounds for volatility forecast with no surprises Robust application for strategies with volatility targeting and time series normalization Volatility Normalized Returns for HY bonds ETF 3 0-3 -6-9 -12-15 Markov Chain CloseClose MarkovChain LowerQuantile MarkovChain UpperQuantile CloseClose LowerQuantile 17
Top-3 models for High Yield Bonds ETF using the normality test annually Use past rolling window of 3 year for one step forecast evaluation Each model is numbered (1,2, ) Stable ranks for Markov chain (31-32) and GARCH models (21-30) Top 3 Estimators with Normality fit for Volatility Normalized Returns for HY bonds ETF Top - 1 Top - 2 Top - 3 32 31 32 32 32 32 32 28 28 28 26 26 26 22 22 22 28 26 26 24 24 26 24 26 26 24 24 31-Dec-2009 31-Dec-2010 30-Dec-2011 31-Dec-2012 31-Dec-2013 31-Dec-2014 31-Dec-2015 30-Dec-2016 29-Dec-2017 18
Top-3 models for the S&P 500 index using normality test in walk-forward analysis annually Markov Chain models (31,32) are frequently on the top Intraday estimators (1-10) are also reliable while being least complex Top 3 Estimators with Normality fit for Volatility Normalized Returns for S&P 500 index Top - 1 Top - 2 Top - 3 32 32 32 32 32 32 32 31 31 31 31 31 31 31 31 30 30 29 29 28 28 28 28 27 27 26 26 25 24 24 24 23 23 23 23 22 22 22 21 21 18 16 16 14 13 12 12 11 10 10 9 8 8 8 6 6 4 4 4 2 1 1 1 1 1 1 19
Quantitative Strategies have changing profile in different market regimes Apply the quantile regression of returns on the strategy vs returns on the benchmark Three regimes: bear, normal, and bull Example using CBOE Put index selling at-the-money put options on the S&P 500 index 20% 10% Quarterly returns on Short Put Index vs S&P 500 index: 1986-2018 Bear Normal y = 1.06x + 0.05 y = 0.43x + 0.02 Bull y = 0.24x + 0.04 Y= Return on Put Index 0% -10% -20% -30% -30% -20% -10% 0% 10% 20% 30% X= Return on the S&P 500 index 20
Risk profile of HFR Bank Systematic Risk Premia Multi-Asset Index vs SG Trend-following CTAs Bank Risk Premia Index is short 3 leveraged put and long 5 leveraged call Trend-following CTAs replicate protection for bear regimes with overall positive performance The difference between amateur and professional applications of ML methods 80% 60% 40% Quarterly returns on HFR Bank Systematic Risk Premia Multi-Asset Index vs S&P 500 index: 2007-2018 Bear Normal Bull y = 2.90x + 0.18 y = 1.46x -0.01 y = 5.30x -0.47 80% 60% 40% Quarterly returns on SG Trend-Following CTAs vs S&P 500 index 2000-2018 Bear Normal Bull y = -0.88x -0.04 y = 0.13x -0.00 y = -1.13x + 0.16 Y= Return on HFR Index 20% 0% -20% -40% Y= Return on SG Index 20% 0% -20% -40% -60% -60% -80% -40% -30% -20% -10% 0% 10% 20% 30% X= Return on the S&P 500 index -80% -30% -20% -10% 0% 10% 20% X= Return on the S&P 500 index 21
Conclusions: Machine Learning for Quant Strategies Machine/Statistical learning models are as good as people behind them Nested approach for strategy design to balance between complexity and approximation & estimation errors Understanding of how the strategy behaves in different market regimes Models adaptation to different regimes: no free or fixed parameters 22
Disclaimer All statements in this presentation are the author personal views. The information and opinions contained herein have been compiled or arrived at in good faith based upon information obtained from sources believed to be reliable. However, such information has not been independently verified and no guarantee, representation or warranty, express or implied, is made as to its accuracy, completeness or correctness. All such information and opinions are subject to change without notice. Descriptions of entities and securities mentioned herein are not intended to be complete. This document is for information purposes only. This document is not, and should not be construed as, an offer, or solicitation of an offer, to buy or sell any securities or other financial instruments. Past performance is not necessarily an indication for future results. This document may not be reproduced, distributed, published or delivered to any other party for any purposes. Investments in Alternative Investment Strategies are suitable only for sophisticated investors who fully understand and are willing to assume the risks involved. Alternative Investments by their nature involve a substantial degree of risk and performance may be volatile. 23