单指标模型的异方差检验的渐近性质  

Asymptotic Property of Testing for Heteroscedasticity in Single Index Models

作　　者：张霞峰[1] 朱仲义[1] 

机构地区：[1]华东师范大学统计系,上海20062

出　　处：《工程数学学报》2007年第2期292-302,共11页Chinese Journal of Engineering Mathematics

基　　金：国家自然科学基金(10371042).

摘　　要：在回归分析中,通常假定方差齐性。在参数和非参数回归模型中,关于方差齐性的检验问题都有很多的研究。本文利用P-样条方法,研究了单指标模型的异方差问题、一阶自回归问题,给出了异方差问题、一阶自回归问题Score检验统计量的大样本性质。The assumption of homoscedasticity is commonly concerned in regression analysis. In parametric and nonparametric regression models, the problem of testing heteroscedasticity, autocor- relation in the linear models has been discussed. In this paper, the testing for heteroscedasticity and autocorrelation is researched for single index models. The score tests are obtained, respectively, under two different occasions. Moreover, we establish the asymptotic property of the test. Simulations show the feasibility and validity of our methods.

关 键 词：X^2分布 异方差 一阶自回归 SCORE检验 单指标模型 P-样条 

分 类 号：O212.2[理学—概率论与数理统计]