Online Program

Saturday, February 23
PS3 Poster Session 3 & Continental Breakfast Sat, Feb 23, 7:30 AM - 9:00 AM
Napoleon Ballroom

Markov Tree Options Pricing: Theory and Empirical Evidence (302591)

Harish Subrahmanya Bhat, University of California, Merced 
*Nitesh Kumar, University of California, Merced  

Keywords: options pricing, empirical testing, delta-hedging, parameter estimation

Option pricing inaccuracies yield incorrect hedges and poor insurance against market risk, which in turn have been linked to systemic banking failure. Standard approaches to fix these inaccuracies are problematic: the practitioner's Black-Scholes model's assumptions conflict with its usage, while Heston's stochastic volatility model leads to computationally intensive parameter estimation. To approach fidelity with the market while avoiding these problems, we propose the Markov Tree (MT) model. Large-scale empirical testing of the MT model versus competing models shows its accuracy with respect to out-of-sample pricing, hedging, and model misspecification. These tests utilize 3 years of Euronext option data stored and analyzed using an efficient, automated MySQL, Python, and R workflow. The MT model's accuracy stems from accounting for short-term memory in the underlying process while requiring only two more parameters than Black-Scholes. We devise a gradient-based method to estimate MT model parameters. Combined with a normal mixture approximation to the MT model's exact option price, this comprises a fast and accurate solution of the pricing problem for European options.