Bates (1996) Stochastic Volatility Jump-Diffusion Lab

Model David S. Bates' (1996) unified derivatives pricing framework blending Heston mean-reverting stochastic variance with Merton compound Poisson jumps. Price European options, generate implied volatility surfaces, and evaluate risk-neutral Greeks.

Bates Call
$0.00
European Call
Bates Put
$0.00
European Put
Jump Contribution
+$0.00
vs Pure Heston
ATM Implied Vol
0.00%
Black-Scholes IV
Total Volatility
0.00%
√(θ + λVar(J))
Feller Ratio
0.00
Satisfied
Option & Underlying Parameters
Heston (1993) Stochastic Volatility
Negative ρ creates standard equity downside volatility skews.
Merton (1976) Compound Jumps
Discrete Poisson shocks capture flash crash risks & steep short-tenor skews.
Quantitative Derivatives Visualizer
Implied Volatility Smile / Skew: Blue curve charts Bates (1996) implied volatility across strikes K ∈ [0.7 S0, 1.3 S0]. Dashed orange curve illustrates pure Heston continuous stochastic volatility (λ=0), dashed green curve shows Merton jump-diffusion, and gray dotted line shows flat Black-Scholes.
Gil-Pelaez Characteristic Function Fourier Inversion (400 Midpoint quadrature nodes).
Institutional Model Decomposition
Model Formulation Call Price Put Price Notes
Bates (1996) Combined $0.00 $0.00 Stochastic Vol + Poisson Jumps
Pure Heston (1993) $0.00 $0.00 Continuous diffusion only (λ = 0)
Pure Merton (1976) $0.00 - Constant vol + jumps (σv = 0)
Black-Scholes (1973) $0.00 - Gaussian baseline (σ = √θ)
Jump Premium (ΔC) +$0.00 - Bates Call − Heston Call
Jump Compensator (κ̄) 0.00% - E[eJ − 1] drift correction
Gil-Pelaez Inversion P1, P2 P1: 0.0000 | P2: 0.0000 Share & Risk-Neutral Probs
Put-Call Parity Error < $0.000001 Exact Martingale Match
Risk-Neutral Sensitivity Greeks
Delta (Δ) 0.0000
∂C / ∂S0 — Share hedge ratio
Gamma (Γ) 0.00000
∂²C / ∂S0² — Convexity curvature
Vega (Ν) $0.00
∂C / ∂√v0 — Dollar change per 1% vol
Theta (Θ) $0.00
∂C / ∂t — 1-day calendar decay
Rho (Ρ) $0.00
∂C / ∂r — Dollar change per 1% interest rate shift
Moneyness & Implied Volatility Smile Table
Strike (Moneyness) Bates Call Bates Put Bates IV (%) Pure Heston IV (%) Jump Spread
Sensitivity 1: Correlation ρ vs. Jump Intensity λ

Evaluates 90% OTM Put Implied Volatility (%). Demonstrates how downward jumps magnify left-tail crash protection pricing under negative spot-volatility correlation.

Sensitivity 2: Strike K vs. Expiry Tenor T

Evaluates European Call Option Price ($). Highlights the term structure of option value across moneyness and expiration horizons.

Theoretical Foundations: The Bates (1996) Model

In financial derivatives pricing, neither the Black-Scholes (1973) model with constant volatility nor the Heston (1993) continuous stochastic volatility model can fully explain real-world option prices across all maturities. While Heston successfully captures the long-dated implied volatility smile and skew through mean-reverting variance and spot-volatility correlation \(\rho\), it fails to generate the steep skew observed in short-dated equity options (\(T < 3\) months) without unrealistically high volatility-of-volatility \(\sigma_v\).

To resolve this puzzle, David S. Bates (1996) introduced the Stochastic Volatility Jump-Diffusion (SVJ) model. Under the risk-neutral pricing measure \(\mathbb{Q}\), the underlying asset price \(S_t\) and its instantaneous variance \(v_t\) follow the coupled bivariate stochastic differential equations:

$$\frac{dS_t}{S_t} = (r - q - \lambda \bar{k}) dt + \sqrt{v_t} dW_t^S + (e^J - 1) dq_t$$ $$dv_t = \kappa (\theta - v_t) dt + \sigma_v \sqrt{v_t} dW_t^v$$ $$d\langle W^S, W^v \rangle_t = \rho dt$$

Here:

Multiplicative Characteristic Function Factorization

Because the continuous Brownian motions \((W_t^S, W_t^v)\) and the Poisson jump process \(q_t\) are stochastically independent, the characteristic function of the terminal log-price factors multiplicatively:

$$\phi_{\text{Bates}}(u; T) = \phi_{\text{Heston}}(u; T) \times \exp\left( \lambda T \left( e^{i u \mu_J - \frac{1}{2} u^2 \sigma_J^2} - 1 - i u \bar{k} \right) \right)$$

European call and put options are then evaluated with high numerical precision via Gil-Pelaez (1951) Fourier inversion:

$$C(S_0, K, T) = S_0 e^{-q T} P_1 - K e^{-r T} P_2$$ $$P_1 = \frac{1}{2} + \frac{1}{\pi} \int_0^\infty \text{Re}\left[ \frac{e^{-i u \ln K} \phi(u - i)}{i u \phi(-i)} \right] du, \quad P_2 = \frac{1}{2} + \frac{1}{\pi} \int_0^\infty \text{Re}\left[ \frac{e^{-i u \ln K} \phi(u)}{i u} \right] du$$

Bates Jump-Diffusion Mastery Quiz

Test your quantitative derivatives understanding of stochastic volatility and jump dynamics.

1. What two classical models does the Bates (1996) model synthesize into a single framework?

2. Why does the pure Heston model struggle to fit the implied volatility skew for short-dated options?

3. Why can the Bates characteristic function be written as the product of Heston's and Merton's transforms?

4. What is the fundamental purpose of the jump compensator k-bar in the drift term?

Frequently Asked Questions

The Bates (1996) model is an institutional continuous-time option pricing framework combining Steven L. Heston's (1993) mean-reverting Cox-Ingersoll-Ross (CIR) stochastic variance process with Robert C. Merton's (1976) compound Poisson lognormal jumps. This hybrid formulation allows traders to capture both persistent long-term volatility clustering and severe short-term implied volatility skews.

Continuous Brownian diffusion requires time to diffuse. As expiration approaches zero (T -> 0), diffusion variance vanishes, forcing the Heston model to produce increasingly flat implied volatility smiles unless the volatility-of-volatility parameter is set to unrealistically extreme levels. Merton-style discrete jumps solve this by allowing immediate finite price leaps regardless of time to expiry.

While Merton's jump-diffusion captures short-dated fat tails and crashes, its jump effect decays rapidly across longer maturities, leaving long-dated implied volatility surfaces flat. Bates incorporates continuous stochastic variance dynamics, enabling the volatility smile and skew to persist naturally across 1-year to 5-year tenors.

Because the continuous Brownian motion driving asset price and volatility is independent of the compound Poisson jump process, the joint characteristic function factors multiplicatively: phi_Bates(u) = phi_Heston(u) * phi_Merton(u). European call and put prices are then solved in milliseconds via Gil-Pelaez Fourier inversion integration.

The jump compensator k-bar = E[e^J - 1] = exp(mu_J + 0.5 * sigma_J^2) - 1 represents the expected percentage return generated by an individual jump. Subtracting lambda * k-bar from the continuous drift under the risk-neutral measure ensures that the discounted underlying asset price remains a strict martingale: E[S_T] = S_0 * exp((r - q) * T).
Related Quantitative Derivatives & Risk Labs
Heston Stochastic Volatility Model Lab

Steven L. Heston (1993) continuous-time bivariate stochastic volatility diffusion and Fourier option pricing.

Merton Jump-Diffusion Option Pricing Lab

Robert C. Merton (1976) compound Poisson jump-diffusion pricing and crash jump risk modeling.

Kou Jump-Diffusion Option Pricing Lab

Steven G. Kou (2002) double exponential jump-diffusion model, asymmetric decays, and memoryless overshoot.

Variance Gamma (VG) Option Pricing Lab

Madan-Carr-Chang pure jump Lévy process, stochastic operational time subordination, and Fourier pricing.

SABR Stochastic Volatility Model Lab

Hagan et al. (2002) asymptotic expansion for interest rate swaptions and volatility smiles.

CEV Model Option Pricing Lab

John Cox (1975) Constant Elasticity of Variance local volatility diffusion, leverage skew, and default absorption.

All Simulation Resources & Curriculum

Complete master curriculum directory indexing all quantitative business, finance, and risk simulations.