摘要
在回归模型中,对一类因变量函数的条件期望方程的附加信息,我们提出了基于极大经验似然方法的局部线性点估计,在一定条件下证明了这些估计的相合性和渐近正态性,而且估计的方差小于通常不带附加信息核估计的方差.模拟结果也显示了估计的优良性.
For a class of conditional expectation equation of dependent variable functions in a regression model, we propose the local linear estimators of regression function and its derivative based on the empirical likelihood method to utilize efficiently auxiliary information. It is shown that the proposed estimators have consistency and asymptotic normality with asymptotic variances less than those of the usual kernel estimators. Simulation study shows our estimators have good performance.
出处
《应用概率统计》
CSCD
北大核心
2004年第2期161-170,共10页
Chinese Journal of Applied Probability and Statistics
关键词
渐近正态性
附加信息
相合性
经验似然
局部线性
Asymptotic normality, auxiliary information, consistency, empirical likelihood, local linear.