作者: Xu-Guo Ye , Jin-Guan Lin , Yan-Yong Zhao , Hong-Xia Hao
DOI: 10.1016/J.JEMPFIN.2015.05.001
关键词:
摘要: Abstract In this article, we develop a two-step estimation procedure for the volatility function in diffusion models. We firstly estimate series at sampling time points based on high-frequency data. Then, estimator can be obtained by using kernel smoothing method. The resulting estimators are presented data, and shown to consistent asymptotically normal. also consider boundary issues then propose two methods handle them. asymptotic normality of boundary-corrected is established under some suitable conditions. proposed illustrated Monte Carlo simulations real