Nonparametric estimation in models with Lévy type jumps and stochastic volatility
We introduce a nonparametric estimator of the volatility function in univariate processes with Lévy type jumps and stochastic volatility when we observe the state variable at discrete times. Our results rely on the fact that it is possible to recognize the discontinuous part of the state variable from those squared increments between observations exceeding a suitable threshold. We discuss the implementation of the estimator with high-frequency data