Merton jump-diffusion model

Merton jump-diffusion model The Merton jump-diffusion model is a stochastic process for asset prices introduced by Robert C. Merton in 1976 that augments continuous Brownian motion with discontinuous jumps modeled by a compound Poisson process. It blends ideas from Bachelier, Black–Scholes model, Paul Samuelson, Stephen Ross, Fischer Black, Myron Scholes, and Louis Bachelier traditions to account for empirical phenomena observed in markets such as sudden price moves and leptokurtic returns. The model influenced later developments by John Hull, Mark Rubinstein, Emanuel Derman, Nassim Nicholas Taleb, and researchers at institutions like Princeton University, Massachusetts Institute of Technology, Harvard University, University of Chicago, and London School of Economics.

Introduction

Merton proposed adding a jump term to the geometric Brownian motion used in the Black–Scholes model to capture rare, large movements in asset prices documented in studies by John Taylor, Clive Granger, and C. W. J. Granger; empirical motivations drew on observations from episodes such as the 1987 stock market crash, the Great Depression, and currency crises like the 1997 Asian financial crisis. The framework combines continuous risk from a Wiener process associated with scholars like Norbert Wiener and discontinuous jumps from a Poisson process studied by Simeon Poisson and further developed in applied contexts by Andrey Kolmogorov, Paul Lévy, and Kiyoshi Itô. It addresses limitations highlighted in critiques by Eugene Fama and later risk analyses by Robert Engle, Clive Granger, and Robert J. Shiller.

Model Definition

The model specifies an asset price S_t satisfying a stochastic differential equation with drift, diffusion, and jump components influenced by concepts from Kiyoshi Itô and jump processes associated with Henri Poincaré and Andrei Kolmogorov. Log-returns combine a Brownian increment tied to Norbert Wiener and a jump term driven by a compound Poisson process parameterized by intensity λ (linked conceptually to work by A. A. Markov and Andrey Kolmogorov) and jump-size distribution commonly assumed log-normal as in models influenced by Benoit Mandelbrot and Eugene Fama. The model uses parameters estimated in line with estimation theory advanced by Jerzy Neyman, Egon Pearson, Ronald Fisher, and time-series methods refined by Clive Granger and Robert Engle.

Option Pricing and Analytical Solutions

Merton derived semi-closed form option pricing formulas extending the Black–Scholes model by summing over Poisson-weighted Black–Scholes prices; this approach resonates with series expansions used in mathematical physics by Poincaré and renewal arguments studied by William Feller. Practical computation draws on numerical techniques from Alan Turing, John von Neumann, and Richard Bellman and on Monte Carlo methods popularized by Stanislaw Ulam and Nicholas Metropolis. Extensions and comparisons involve models by David S. Bates, jump-diffusion analyses by Robert Engle collaborators, and affine jump-diffusion frameworks connected to work by Darrell Duffie and Damir Filipović. Option Greeks and hedging strategies relate to sensitivity analysis traditions in Paul Samuelson and risk management practices at institutions like Goldman Sachs, J.P. Morgan, and Morgan Stanley.

Calibration and Estimation

Calibration to market data employs maximum likelihood estimation and characteristic function methods influenced by Marc Kac and Carl Friedrich Gauss; practitioners use implied volatility surface fitting techniques paralleling research by Louis Bachelier and later empirical work by Mark Rubinstein, Emanuel Derman, and Ivan Štěpán. Econometric procedures borrow from generalized method of moments developed by Lars Peter Hansen and state-space filtering approaches derived from the Kalman filter by Rudolf E. Kálmán; particle filtering and sequential Monte Carlo algorithms credit work by Gordon, Salmond and Smith and Alan Doucet. Calibration challenges were highlighted after crises investigated by Financial Stability Board and scholars like Robert Shiller and Nassim Nicholas Taleb, prompting robust estimation techniques used at central banks such as the Federal Reserve System, Bank of England, and European Central Bank.

Applications and Extensions

The model underpins pricing of corporate debt and credit derivatives in frameworks influenced by Merton's earlier structural credit model, and it informed jump-extended stochastic volatility models developed by Steven Heston, Jim Gatheral, and Tomasz R. Bielecki. Applications span equity derivatives trading at firms like Citigroup and Deutsche Bank, risk analyses in portfolio management at BlackRock and Vanguard Group, and scenario modeling used by regulators including Securities and Exchange Commission and Basel Committee on Banking Supervision. Academic extensions include regime-switching jumps related to James Hamilton's work, Lévy process generalizations from Ken-iti Sato, infinite-activity models by Ole E. Barndorff-Nielsen, and mixed jump-diffusion frameworks explored at INSEAD, Wharton School, and Stanford Graduate School of Business.

Category:Mathematical finance