误差为MA(∞)序列的半参数回归模型小波估计的渐近正态性  

Asymptotic Normality of Wavelet Estimators in Semiparametric Regression Models With MA(∞) Time Series Errors

作　　者：胡宏昌[1] 

机构地区：[1]湖北师范学院数学与统计学院,黄石湖北435002

出　　处：《数学进展》2013年第4期551-562,共12页Advances in Mathematics（China）

基　　金：国家自然科学基金(No.11071022);湖北省教育厅青年项目(No.Q20122203)

摘　　要：用小波估计研究误差为MA(∞)序列的半参数回归模型,在比较弱的条件下得到了自协方差函数及自相关函数估计量的渐近正态性,并且用小波估计法建立了英国烈酒消费模型,说明了该法的有效性.In the paper, we consider semiparametric regression models with MA（∞） time series errors by wavelet method. Under some relatively weak conditions, we establish asymptotic normality of estimators of the autocovariance and the autocorrelation functions. In order to illustrate the validity and advantages of the wavelet method, we investigate a model of annual consumption of spirits in the United Kingdom by the method.

关 键 词：半参数回归模型 MA(∞)序列 小波估计 渐近正态性 

分 类 号：O212.1[理学—概率论与数理统计]