A reexamination of diffusion estimators with applications to financial model validation

摘要: Time-homogeneous diffusion models have been widely used for describing the stochastic dynamics of underlying economic variables. Recently, Stanton proposed drift and estimators based on a higher-order approximation scheme kernel regression method. He claimed that “higher order approximations must outperform lower approximations” concluded nonlinearity in instantaneous return function short-term interest rates. To examine impact approximations, we develop general explicit formulas asymptotic behavior both estimators. We show these will reduce numerical errors biases, but their variances escalate nearly exponentially with approximation. Simulation studies also confirm our results. This variance inflation problem arises not only from nonparametric fitting, parametric fitting. Stanton's work postulates interesting qu...