股票价格稳定性的非参数平滑估计  

Nonparametric Smoothed Estimation of Stability of Stock Price

作　　者：王俏[1] 王文举[1] 

机构地区：[1]首都经济贸易大学经济学院,北京100070

出　　处：《首都经济贸易大学学报》2015年第6期22-29,共8页Journal of Capital University of Economics and Business

基　　金：国家社会科学基金重大项目"中国碳市场成熟度;市场机制完善及环境监管政策研究"(14ZDA072);北京市属高等学校高层次人才引进与培养计划"长城学者"资助项目"碳排放与博弈计量研究"(CIT&TCD20140321)

摘　　要：股票的优质性主要由股票价格的稳定性来决定。通过建立条件风险函数非参数平滑估计的估计方法,证明条件风险函数平滑后的非参数估计值具有一致性和渐进正态性,并将此方法扩展到了截尾数据中。通过蒙特卡罗模拟表明条件风险函数平滑后的非参数估计值在非截尾数据以及截尾数据中的有限样本行为都要优于非平滑的估计值。实证分析中用截尾数据条件风险函数非参数平滑估计方法非参数估计中国14大行业股票价格的稳定性,发现房地产行业的股票价格最不稳定,建筑材料行业的股票价格表现最稳定。The quality of a stock is mainly determined by its stability. Firstly,a nonparametric smoothed estimation of conditional hazard function is proposed. The consistency and asymptotic normality of the estimator and extend the kernel method to censored data are proved. The simulation study indicates that the smoothed estimator of conditional hazard function without or with censored data performs better than the non-smoothed estimator in finite samples. Secondly,the stability of 14 industries＇ stock price is estimated with the smoothed method. The estimation result shows that the stock price of real estate is the least stable and the stock price of construction is the most stable.

关 键 词：条件风险函数 非参数平滑估计 截尾数据 

分 类 号：O212[理学—概率论与数理统计]