A comparative simulation study for estimating diffusion coefficient Original Research Article

An efficient methodology of estimation of parameters in the diffusion coefficient of the stochastic differential equation (SDE) is presented in this work. The methodology is based on the concept of quadratic variation of a stochastic process and on some classical numerical tools such as spline quadrature and bisection method. The robustness of the procedure is tested with respect to the different discretization schemes for SDE and to different form and values of drift and diffusion coefficients. The results reveal good performances in all the different situations taken into account.