Threshold estimation of Markov models with jumps and interest rate modeling

摘要: We reconstruct the level-dependent diffusion coefficient of a univariate semimartingale with jumps which is observed discretely. The consistency and asymptotic normality our estimator are provided in presence both finite infinite activity (finite variation) jumps. Our results rely on kernel estimation, using properties local time data generating process, fact that it possible to disentangle discontinuous part state variable through those squared increments between observations not exceeding suitable threshold function. also drift jump intensity coefficients when they have activity, consistent asymptotically normal estimators. Simulated experiments show newly proposed estimators perform better samples than alternative estimators, this allows us reexamine estimation model for short term interest rate, we find fewer more variance due previous studies.