Back to the past: option pricing using realized volatility



Similar documents
Option Valuation Using Intraday Data

Implied Volatility Surface

Modeling the Implied Volatility Surface. Jim Gatheral Stanford Financial Mathematics Seminar February 28, 2003

THE ECONOMIC VALUE OF TRADING WITH REALIZED VOLATILITY

Stock Price Dynamics, Dividends and Option Prices with Volatility Feedback

Private Equity Fund Valuation and Systematic Risk

ARMA, GARCH and Related Option Pricing Method

Affine-structure models and the pricing of energy commodity derivatives

Arbitrage-Free Pricing Models

Investors and Central Bank s Uncertainty Embedded in Index Options On-Line Appendix

Stochastic Skew in Currency Options

How To Price Garch

Introduction Pricing Effects Greeks Summary. Vol Target Options. Rob Coles. February 7, 2014

Risk and return (1) Class 9 Financial Management,

Lecture 1: Asset pricing and the equity premium puzzle

Study on the Volatility Smile of EUR/USD Currency Options and Trading Strategies

Using the SABR Model

On Marginal Effects in Semiparametric Censored Regression Models

The term structure of equity option implied volatility

IL GOES OCAL A TWO-FACTOR LOCAL VOLATILITY MODEL FOR OIL AND OTHER COMMODITIES 15 // MAY // 2014

VI. Real Business Cycles Models

Charles University, Faculty of Mathematics and Physics, Prague, Czech Republic.

Credit Risk Models: An Overview

A Regime-Switching Model for Electricity Spot Prices. Gero Schindlmayr EnBW Trading GmbH

Retrieving Risk Neutral Moments and Expected Quadratic Variation from Option Prices

Estimating Volatility

Dynamic Jump Intensities and Risk Premiums in Crude Oil Futures and Options Markets

Implied Volatility of Leveraged ETF Options

The Effect of Housing on Portfolio Choice. July 2009

Implied volatility surface

Volatility Forecasting I: GARCH Models

No-Arbitrage Condition of Option Implied Volatility and Bandwidth Selection

Delta-hedging Vega Risk?

Heterogeneous Beliefs and The Option-implied Volatility Smile

Volatility Forecasting Performance: Evaluation of GARCH type volatility models on Nordic equity indices

Hedging Illiquid FX Options: An Empirical Analysis of Alternative Hedging Strategies

Estimation of Stochastic Volatility Models with Implied Volatility Indices and Pricing of

Estimating the Degree of Activity of jumps in High Frequency Financial Data. joint with Yacine Aït-Sahalia

Markov modeling of Gas Futures

Caput Derivatives: October 30, 2003

A non-gaussian Ornstein-Uhlenbeck process for electricity spot price modeling and derivatives pricing

Modelling Intraday Volatility in European Bond Market

Generating Random Numbers Variance Reduction Quasi-Monte Carlo. Simulation Methods. Leonid Kogan. MIT, Sloan , Fall 2010

Spatial Statistics Chapter 3 Basics of areal data and areal data modeling

Real-time Volatility Estimation Under Zero Intelligence. Jim Gatheral Celebrating Derivatives ING, Amsterdam June 1, 2006

Expected default frequency

Sensex Realized Volatility Index

Understanding Volatility

Volatility Density Forecasting & Model Averaging Imperial College Business School, London

Investment Statistics: Definitions & Formulas

Volatility Index: VIX vs. GVIX

Jumps and microstructure noise in stock price volatility

Recent Developments of Statistical Application in. Finance. Ruey S. Tsay. Graduate School of Business. The University of Chicago

Valuing Stock Options: The Black-Scholes-Merton Model. Chapter 13

A comparison between different volatility models. Daniel Amsköld

Intraday Trading Invariance. E-Mini S&P 500 Futures Market

(Arbitrage-Free, Practical) Modeling of Term Structures of Government Bond Yields

by Maria Heiden, Berenberg Bank

Chapter 7. Statistical Models of Operational Loss

A constant volatility framework for managing tail risk

Lisa Borland. A multi-timescale statistical feedback model of volatility: Stylized facts and implications for option pricing

GN47: Stochastic Modelling of Economic Risks in Life Insurance

Credit Implied Volatility

Option Pricing with Time Varying Volatility

How to assess the risk of a large portfolio? How to estimate a large covariance matrix?

Options 1 OPTIONS. Introduction

Markov Chain Monte Carlo Simulation Made Simple

RISK MANAGEMENT TOOLS

A constant volatility framework for managing tail risk

The Impact of Target Volatility Strategies on Variable and Indexed Annuities

3. Regression & Exponential Smoothing

Supplement to Call Centers with Delay Information: Models and Insights


Hedging Options In The Incomplete Market With Stochastic Volatility. Rituparna Sen Sunday, Nov 15

Markups and Firm-Level Export Status: Appendix

Monte Carlo-based statistical methods (MASM11/FMS091)

INDIRECT INFERENCE (prepared for: The New Palgrave Dictionary of Economics, Second Edition)

Does the Spot Curve Contain Information on Future Monetary Policy in Colombia? Por : Juan Manuel Julio Román. No. 463

Understanding the Impact of Weights Constraints in Portfolio Theory

Master of Mathematical Finance: Course Descriptions

The Heston Model. Hui Gong, UCL ucahgon/ May 6, 2014

A Log-Robust Optimization Approach to Portfolio Management

Online Appendix for Demand for Crash Insurance, Intermediary Constraints, and Risk Premia in Financial Markets

VDAX -NEW. Deutsche Börse AG Frankfurt am Main, April 2005

Transcription:

Back to the past: option pricing using realized volatility F. Corsi N. Fusari D. La Vecchia Swiss Finance Institute and University of Siena Swiss Finance Institute, University of Lugano University of Lugano 12th October, Kellogg School of Management

Outline Introduction Motivation Previous work Stylized facts under the historical measure Model properties Risk-neutral measure Database description Conclusions

Introduction Motivation Outline Introduction Motivation Previous work Stylized facts under the historical measure Model properties Risk-neutral measure Database description Conclusions

Introduction Motivation Motivation High frequency data are nowdays wide available and easily tractable; considerable interest has been recently devoted to the use of high frequency (HF) data for measuring and forcasting volatility; volatility is a key ingredient of every option pricing model; only few papers try to employ high frequency data to price options or other derivatives. GOAL: definition of a discrete-time model for option pricing using Realized Volatility (RV) as a proxy for the unobservable underlying volatility.

Introduction Previous work Outline Introduction Motivation Previous work Stylized facts under the historical measure Model properties Risk-neutral measure Database description Conclusions

Introduction Previous work Previous work Realized volatility measure and forcast: Andersen et al. (21) and (23), Barndorff-Nielsen and Shephard (21) (22a) (22b) and (25), Comte and Renault (1998); Option pricing using RV: Stentoft (28).

Stylized facts Outline Introduction Motivation Previous work Stylized facts under the historical measure Model properties Risk-neutral measure Database description Conclusions

Stylized facts General idea A well established result in the financial econometrics literature is that, daily log returns do not have a Gaussian distribution: leptokurtic heavy-tailed distributions. Clark (1973) and Ane and Geman (2): for an underlying semi-martingale process, rescaling the log returns by an appropriate measure of the market activity allows to recover the standard Gaussian distribution. Here we focus on Realized Volatility as a measure of market activity (which is per se not observable).

under the historical measure Outline Introduction Motivation Previous work Stylized facts under the historical measure Model properties Risk-neutral measure Database description Conclusions

under the historical measure Model set-up Methodological approach: We cast our model in the general framework proposed by Bertholon, Monfort and Pegoraro (28). Specifically, in pricing derivatives, one can follow three different approaches: direct modeling (P and SDF); Risk-neutral constrained modeling (Q and P); back modeling (Q and SDF). Here we follow the first alternative (direct modeling).

under the historical measure under the historical measure Log-returns We assume the following dynamics for the log returns: ( ) St+1 ln := y t+1 = µ t+1 + RV t+1 ɛ t+1, (1) S t ɛ t+1 RV t+1 N(, 1). In our notation, S t+1, y t+1, and RV t+1 are respectively the price, the log returns and the RV at time t + 1. For the drift of the log returns (i.e. µ t+1 ) we propose the following specification: µ t+1 = r + ( γ 1/2) RV t+1.

under the historical measure under the historical measure (cont d) Realized volatility dynamics Which process to describe the RV dynamics? - Empirically we observe that the main feature of the RV is the strong persistence. = Therefore we follow Corsi (29) and we model the conditional mean of the RV (given its past values) using the conditional expected value of an HAR (Heterogeneous AutoRegressive) process.

under the historical measure under the historical measure (cont d) Realized volatility dynamics We model the RV as an Auto Regressive Gamma process (ARG) (see Gourieroux and Jasiack 26). This implies that: RV t+1 F t γ(δ, β (RV t, L t ), c) δ and c describe, respectively, the shape and the scale of the distribution, whereas β (RV t, L t ) is the location parameter given by: β (RV t, L t ) = β 1 RV t + β 2 ( 4 RV t i ) + β 3 ( i=1 21 i=5 RV t i ) + ΛL t. (2) We label this model Heterogeneous AutoRegressive Gamma (HARG) process.

Model properties Outline Introduction Motivation Previous work Stylized facts under the historical measure Model properties Risk-neutral measure Database description Conclusions

Model properties Model simulation 1 S&P5 log returns 5 5 1 199 1992 1995 1997 2 22 25 27 Simulated log returns 1 5 5 1 199 1992 1995 1997 2 22 25 27 Figure: Log-returns. Top panel: S&P 5 log-returns from January 1st 199 to December 31th 27 (4218 observations). Bottom panel: log-returns simulated path from HARG(3).

Model properties Variance term structure Term structure of GARCH(1,1) model 2 High h t+1 Term structure of component model 2 2 Term structure of HARG model Low h t+1 Normalized varance 1.5 1 Normalized varance 1.5 1 Normalized varance 1.5 1.5 5 1 15 2 25 Horizon in days.5 5 1 15 2 25 Horizon in days.5 5 1 15 2 25 Horizon in days

Model properties Skewness-Kurtosis term structure Skewness Skewness Skewness.5 1 5 1 15 2 25 Days to maturity Component GARCH skewness.5 GARCH(1,1) skewness 1 5 1 15 2 25 Days to maturity HARG skewness Simulated.5 Fitted 1 1.5 5 1 15 2 25 Days to maturity Excess kurtosis Excess kurtosis Excess kurtosis 1.5 5 1 15 2 25 Days to maturity Component GARCH excess kurtosis 1.5 GARCH(1,1) excess kurtosis 5 1 15 2 25 Days to maturity HARG excess kurtosis 1 Simulated Fitted 5 5 1 15 2 25 Days to maturity

Model properties Volatility and Correlation Daily standard deviation Daily standard deviation Daily standard deviation x 1 5 GARCH(1,1): Stdev of h 3 2 1 199 2 21 x 1 5 Component GARCH: Stdev of h 3 2 1 199 2 21 4 x HARG: Stdev of RV 1 4 2 199 2 21 Daily correlation Daily correlation Daily correlation.5 GARCH(1,1): Corr(R,h) 1 199 2 21.5 Component GARCH: Corr(R,h) 1 199 2 21.2.4.6 HARG: Corr(R,h).8 199 2 21

Risk-neutral measure Outline Introduction Motivation Previous work Stylized facts under the historical measure Model properties Risk-neutral measure Database description Conclusions

Risk-neutral measure The stochastic discount factor The HARG(3) is an affine process under P. In order to maintain the analytical tractability of the model under Q we set up an exponential affine stochastic discount factor (SDF): M t,t+1 = exp( ν ν 1RV t+1 ν 2RV d t ν 3RV w t 1 ν 4RV m t 5 ν 5y t+1 ν 6L t). This SDF is compatible with the no-arbitrage conditions (parameter restrictions - in the paper).

Risk-neutral measure The (resulting) model under the risk neutral measure Under Q, the log returns follow a discrete-time stochastic model, with risk premium γ = 1/2. The RV is an HARG(3) process, featuring a transition density given by a non central gamma, with parameters ρ, δ, µ. with λ = ν 1 + γ2 2 1 8. ρ = cβ (1 + cλ) 2, (3) δ = δ, (4) c = c 1 + cλ, (5)

Database description Dataset We employ a large dataset of options taken from OptionMetrics from 1st January 1996 to 27th December 24. We consider only OTM Call-Put options with volatility less then 7%, maturity between 1 and 365 days and whose price is not less than.5 dollars.

Outline Introduction Motivation Previous work Stylized facts under the historical measure Model properties Risk-neutral measure Database description Conclusions

Competing models The direct modeling approach allows to estimate the HARG(3) model under the historical measure using the time series of log-returns and realized volatility. Just one parameter (ν 1 ) has to be calibrated. From a practical point of view we use the most liquid ATM Call and Put (minimizing the RMSE IV ). Competitor models: HN GARCH(1,1), Component GARCH.

Option pricing performace - Global HARG HN GARCH(1,1) Component GARCH RMSE p 4.2618.578.6573 RMSE IV 5.9838.6469.8126 Table: Option pricing performance.

RMSE dynamics IV RMSE (%) 16 14 12 1 8 6 4 2 IV RMSE dynamics 1996 1998 2 22 24 26

Nonparametric volume surface

HARG vs HN GARCH(1,1) Maturity Moneyness Less then 2 2 to 6 6 to 16 More then 16 < 2.4591.5915.4318.3236 [ 2, 1).7771.716.5286.4 [ 1,.5).7542.889.6328.4786 [.5,.5).7367.8751.887.6284 [.5, 1).914 1.53 1.1544 1.1529 >= 1.746.7974.6444.9624 Table: HARG(3) vs HN GARCH(1,1).

HARG vs Component GARCH Maturity Moneyness Less then 2 2 to 6 6 to 16 More then 16 < 2.4535.613.522.414 [ 2, 1).7786.7665.657.5318 [ 1,.5).772.8223.844.674 [.5,.5).7864.7857.9132.9217 [.5, 1) 1.449.8144.8944 1.3573 >= 1.7141.94.8419 1.2395 Table: HARG(3) vs Component GARCH.

Option pricing performance - ATM options bias GARCH(1,1) Component GARCH HARG Implied volatility bias.15.1.5.5.1.15 Jan95 Jan Jan5 Implied volatility bias.15.1.5.5.1.15 Jan95 Jan Jan5 Implied volatility bias.15.1.5.5.1.15 Jan95 Jan Jan5

Volatility regimes 5 VIX 45 4 High volatility VIX level (%) 35 3 25 2 15 1 Medium volatility Low volatility 5 199 1992 1995 1997 2 22 25 27

Option pricing performance - ATM options term structure Low implied volatility Medium implied volatility High implied volatility 25 2 15 1 5 1 15 2 25 3 35 Days to maturity 22 2 18 16 5 1 15 2 25 3 35 Days to maturity 3 25 2 15 5 1 15 2 25 3 35 Days to maturity

Option pricing performance - cross-section 6<DTM<2 2<DTM<6 6<DTM<16 16<DTM<365 4 2 Low volatility period 3 2 1 1 2 Moneyness 3 2 1 3 2 1 1 2 Moneyness 3 2 1 3 2 1 1 2 Moneyness 4 2 3 2 1 1 2 Moneyness 6<DTM<2 2<DTM<6 6<DTM<16 16<DTM<365 4 2 Medimu volatility period 3 2 1 1 2 Moneyness 4 2 3 2 1 1 2 Moneyness 4 2 3 2 1 1 2 Moneyness 4 2 3 2 1 1 2 Moneyness 6<DTM<2 2<DTM<6 6<DTM<16 16<DTM<365 4 2 High volatility period 3 2 1 1 2 Moneyness 4 2 3 2 1 1 2 Moneyness 4 2 3 2 1 1 2 Moneyness 4 2 3 2 1 1 2 Moneyness

Option pricing performance - cross-section Implied Volatility.35.3.25.2 Data cross section 1 Month 6 Months 1 Year 3 2 1 1 2 Moneyness Implied Volatility.35.3.25.2 HARG corss section 1 Month 6 Months 1 Year 3 2 1 1 2 Moneyness Implied Volatility.25.24.23.22.21.2.19 CJOW cross section 1 Month 6 Months 1 Year.18 3 2 1 1 2 Moneyness Implied Volatility.35.3.25.2 Models comparison 1 month Data CJOW HARG Implied Volatility.35.3.25.2 Models comparison 6 month Data CJOW HARG Implied Volatility.35.3.25.2 Models comparison 12 month Data CJOW HARG 3 2 1 1 2 Moneyness 3 2 1 1 2 Moneyness 3 2 1 1 2 Moneyness

Option pricing performance - IV surface dynamics Level dynamics Level Term structure.4.3.2.1 1996 1998 2 22 24 26.2 Term structure dynamics.2 1996 1998 2 22 24 26.2 Skew dynamics Skew.1 1996 1998 2 22 24 26

Conclusions Conclusions We develop a stochastic volatility option pricing model that exploits the historical information contained in the HF data. Straightforward estimation. No need to calibrate all the parameters. The HARG(3) outperform competing GARCH models.