） 摘要： 综文的研究对象为带有结构变点的AR(1)模型,其中AR参数β在未知时刻k0发生变化
即模型中有两个AR参数$β1和β2
假设: (I)β1绝对值小于1, β2=β2T=1-c/T; (II)β1=β1T=1-c/T, β2绝对值小于1, 其中c为一固定的常数
此外, 还假设 {εt, t≥1}为来自正态吸引场的独立同分布随机变量,数学期望为0,方差可能不存在
那么我们在本文中讨论了β1和β2的最小二乘估计, 以及具有收缩性质的变点的估计的极限分布
关键词： AR(1)模型
张丹娜
带有变点的AR(1)模型的渐近推断：平稳和近似非平稳情形[EB/OL]
北京：中国科技论文在线
ZHANG Danna 2, （ 1、 Department of Mathematics, Yuquan Campus, Zhejiang University, HangZhou 310027
2、 芝加哥大学统计系,伊利诺伊 60637, 美国
） Abstract： The basic model in this paper is an AR(1) model with a structural break in the AR parameter β atan unknown time k0
That is, there are two AR parameters, β1 and β2, in the model
Suppose: (I) β1 is smaller than one in absolute value, β2=β2T=1-c/T; (II)β1=β1T=1-c/T, β2 is smaller than one in absolute value, with c being a fixed constant, and additionally, let {εt, t≥1} be a sequence of i
i
d
random variables which are in the domain of attraction of thenormal law with zero means and possibly infinite variances
Then the limiting distributions of the least squares estimators of β1 and β2, and the limiting distribution of the break-point estimator for shrinking break are all studied in the present paper under both of the above-mentioned cases
Keywords： AR(1) model; Change point; Domain of attraction of the normal law; Limiting distribution; Least squares estimator 下载PDF阅读器 PDF全文下载： 初稿 （ 140 ） 原始数据 作者简介： Pang Tianxiao,(1979-), Associate Professor
通信联系人： 【收录情况】 中国科技论文在线： 庞天晓
2、 Department of Statistics, University of Chicago, 5734 S
University Ave
, Chicago, Illinois 60637, USA
