摘要
考虑了一类基于样本插值的时齐扩散方程扩散函数的非参数估计程序.在一定的正则条件下,给出的扩散函数估计量是依概率收敛的,并且渐近地符合一个正态分布.通过分析,发现与传统的基于已实现波动率的估计量相比,估计量在精度上有所提高.
In this paper, we consider a nonparametric estimation procedure for diffusion function of homogeneous stochastic differential equations based on sample interpolation. Un- der some certain regularity conditions, the proposed diffusion function estimator is consistent in probability and asymptotically follows a normal distribution. By our theoretical results, the precision of our estimator is better than that of the conventional estimator based on realized volatility.
出处
《数学的实践与认识》
北大核心
2015年第12期183-191,共9页
Mathematics in Practice and Theory
基金
国家自然科学基金(11271189
11201229)
关键词
样本插值
扩散函数
非参数估计
精度
sample interpolation
diffusion function
nonparametric estimation
precision
