摘要
设y_1,…,y_n是在固定点x_1,…,x_n的n个观察值,适合模型y_i=g(x_i)+ε_i,1≤i≤n,这里g是R^1上的未知函数,{ε_i}为零均值、有限方差的i.i.d.序列,设O=x_o≤x_l≤…≤x_(n-1)≤x_n=1我们用作为g^(p)(x)的估计. 本文给出了的渐近分布,其中x^(1),…,x^(m)为(0,1)内的相异点。
Let y1,…,yn be n observations at fixed points x 1,…,xn according to the modelwhere g is an unknown function on R1 and (ct} are i'.i.d.random variables with zero means and finite varians.Assumc that 0 . We useas a estimator of g (x).In this paper, we establish the conditions under which is asymptotically mulitivariatenormal,where x(1),…,x(m)∈(0,1)and x(1)(1)≠x(f) for i≠j, i,j = 1 ,…,m.
出处
《工程数学学报》
CSCD
1991年第1期67-74,共8页
Chinese Journal of Engineering Mathematics
