多元秩-序模型中回归系数的估计  

Estimation of regression coefficient in multivariate rank-order model

作　　者：马建军[1] 徐兴忠[1] 

机构地区：[1]北京理工大学数学系

出　　处：《高校应用数学学报（A辑）》2008年第3期287-294,共8页Applied Mathematics A Journal of Chinese Universities(Ser.A)

摘　　要：通过对多元秩-序模型的研究得到了模型的逆回归性质,基于该性质提出了回归系数的估计方法.当自变量满足线性条件时,不用预先设定扰动项的具体分布便可以得到回归系数方向的估计,并且这个估计与回归系数只相差一个正常数因子.证明了估计是n^(1/2)相合的.模拟结果表明估计有良好的大样本性质.The multivariate rank-order model is investigated and an inverse regression property is found.Based on this property,a method is proposed to estimate the regression coefficient.When the regressors satisfy the linearity condition,the regression coefficient can be estimated up to a positive multiplicative scale without specifying the distribution of errors.The x/^-consistency of the estimator is established.The simulation results show that the estimator performs well when the sample size gets large.

关 键 词：LIMITED DEPENDENT variable模型 多元秩-序模型 逆回归 

分 类 号：O212.1[理学—概率论与数理统计]