A Damped Diffusion Framework for Financial Modeling and Closed-Form Maximum Likelihood Estimation

Posted: 5 Feb 2010

Multiple version iconThere are 2 versions of this paper

Date Written: February 5, 2010

Abstract

Asset price bubbles can arise unintentionally when one uses continuous-time diffusion processes to model financial quantities. We propose a flexible damped diffusion framework that is able to break many types of bubbles and preserve the martingale pricing approach. Damping can be done on either the diffusion or drift function. Oftentimes, certain solutions to the valuation PDE can be ruled out by requiring the solution to be a limit of martingale prices for damped diffusion models. Monte Carlo study shows that with finite time-series length, maximum likelihood estimation often fails to detect the damped diffusion function while fabricates nonlinear drift function. An alternative method based on Aı¨t-Sahalia’s specification test on parametric models is proposed.

Keywords: Damped diffusion, Asset Price Bubbles, Martingale Pricing, Maximum Likelihood Estimation

JEL Classification: C60, G12, G13

Suggested Citation

Li, Minqiang, A Damped Diffusion Framework for Financial Modeling and Closed-Form Maximum Likelihood Estimation (February 5, 2010). Journal of Economic Dynamics and Control, Vol. 34, No. 2, 2010, Available at SSRN: https://ssrn.com/abstract=1548522

Minqiang Li (Contact Author)

Bloomberg LP ( email )

731 Lexington Avenue
New York, NY 10022
United States

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
427
PlumX Metrics