Diffusion on the circle and a stochastic correlation model

Abstract: We propose analytically tractable SDE models for correlation in financial markets. We study diffusions on the circle, namely the Brownian motion on the circle and the von Mises process, and consider these as models for correlation. The von Mises process was proposed in Kent (1975) as a probabilistic justification for the von Mises distribution which is widely used in Circular statistics. The transition density of the von Mises process has been unknown, we identify an approximate analytic transition density for the von Mises process. We discuss the estimation of these diffusion models and a stochastic correlation model in finance. We illustrate the application of the proposed model on real-data of equity-currency pairs.