Parametric estimation in skew-sticky diffusions via local time approximation
par
We consider certain one-dimensional diffusion processes whose dynamics are perturbed by the presence of a point barrier that is either partially reflective (skew) or sticky.
The nature of this barrier is characterized by skewness and stickiness parameters.
We first describe the process and its key properties, then discuss the approximation of local time and the estimation of these parameters from a single trajectory observed at discrete times.
We explore the convergence behavior of the estimators in the presence of nontrivial skewness and/or stickiness, and in particular whether they converge to a mixed Gaussian distribution at the same rate as in the standard Brownian motion case.
This talk is partly based on joint works with A. Anagnostakis (IECL, Metz).