您的位置:首页  > 论文页面

基于均匀分布协变量线性模型的参数估计

发表时间:2013-07-15  浏览量:1270  下载量:533
全部作者: 刘东梅,吴晓萍,赵学靖
作者单位: 兰州大学数学与统计学院
摘 要: 针对区间删失数据的特点,研究当协变量为区间删失型数据时线性模型的参数估计。通过构造区间删失数据下协变量的2种近似方法——离散化方法和大样本正态逼近,用极大似然估计(maximum likelihood estimation, MLE)方法得到了回归参数的估计,并对其结果进行模拟和比较分析。研究结果表明:在协变量为区间删失数据的线性模型中,运用离散化方法能得到未知参数的较高精度的估计,且该方法较传统方法有理论保障;另外,在大样本情形下,采用正态逼近的方法亦能得到各参数间很小误差的估计,结果令人满意。
关 键 词: 应用统计数学;区间删失数据;回归模型;极大似然估计
Title: Parametic estimation of linear model based on uniformly distributed covariate
Author: LIU Dongmei, WU Xiaoping, ZHAO Xuejing
Organization: School of Mathematics & Statistics, Lanzhou University
Abstract: This paper focused on parametric estimation of linear model for interval-censored covariates. By using two approximation methods (discretization method and large sample normal approximation) of the interval-censored covariates, the results of the parametric estimation with the maximum likelihood estimation (MLE) method was obtained. Furthermore, the performance of the proposed methods was illustrated by numerical simulations comparison analysis. The results showed that, in the linear model with an interval censor-data covariates, a higher accuracy of estimation of unknown parameters could be obtained by using the discretization method, and the error was small when comparing with the traditional methods. In addition, a good estimation of parameter could also be obtained in the case of large sample, and the result was satisfactory.
Key words: applied statistical mathematics; interval-censored covariates; regression model; maximum likelihood estimation
发表期数: 2013年7月第13期
引用格式: 刘东梅,吴晓萍,赵学靖. 基于均匀分布协变量线性模型的参数估计[J]. 中国科技论文在线精品论文,2013,6(13):1221-1226.
 
0 评论数 0
暂无评论
友情链接