应用概率统计
應用概率統計
응용개솔통계
CHINESE JOURNAL OF APPLIED PROBABILITY AND STATISTICS
2013年
2期
167-178
,共12页
变系数模型%核估计%置信区间%渐近正态%纵向数据
變繫數模型%覈估計%置信區間%漸近正態%縱嚮數據
변계수모형%핵고계%치신구간%점근정태%종향수거
Varying coefficient model%kernel estimator%confidence region%asymptotic normality%longitudinal data
考虑纵向数据下变系数EV模型,提出了未知系数函数的局部纠偏光滑核估计,在适当条件下,证明了所提出的光滑核估计具有渐近正态性,由此构造回归系数的逐点置信区间.模拟研究了所提出方法的有限样本性质.
攷慮縱嚮數據下變繫數EV模型,提齣瞭未知繫數函數的跼部糾偏光滑覈估計,在適噹條件下,證明瞭所提齣的光滑覈估計具有漸近正態性,由此構造迴歸繫數的逐點置信區間.模擬研究瞭所提齣方法的有限樣本性質.
고필종향수거하변계수EV모형,제출료미지계수함수적국부규편광활핵고계,재괄당조건하,증명료소제출적광활핵고계구유점근정태성,유차구조회귀계수적축점치신구간.모의연구료소제출방법적유한양본성질.
@@@@Varying coefficient EV models with longitudinal data are considered. The local bias-corrected kernel estimators for the unknown coe?cient functions are proposed. It is shown that the proposed estimators are asymptotically normal under some suitable conditions, and hence it can be used to construct the pointwise confidence regions of the coefficient functions. The finite-sample properties of the proposed procedures are studied through a simulation study.