首页 | 本学科首页   官方微博 | 高级检索  
   检索      

多重二元响应Probit模型的渐近有效估计
引用本文:马建军,徐兴忠.多重二元响应Probit模型的渐近有效估计[J].生物数学学报,2008,23(4):677-686.
作者姓名:马建军  徐兴忠
作者单位:1. 沈阳理工大学,理学院,辽宁,沈阳,110168
2. 北京理工大,学数学系,北京,100081
摘    要:针对多重二元响应Probit模型提出了两步估计方法,第一步由边际似然得到参数√n相合的估计,第二步通过一步迭代得到渐近有效估计,由于只需一步迭代,因此在利用模拟方法计算信息阵时,可以增加模拟的次数,从而减少模拟所产生的扰动对估计的影响.

关 键 词:Limited  DEPENDENT  variable模型  多重二元响应模型  PROBIT模型  极大似然估计  渐近有效估计

The Asymptotic Efficient Estimation of the Multi-binary Response Probit Model
MA Jian-jun,XU Xing-zhong.The Asymptotic Efficient Estimation of the Multi-binary Response Probit Model[J].Journal of Biomathematics,2008,23(4):677-686.
Authors:MA Jian-jun  XU Xing-zhong
Institution:MA Jian-jun, XU Xing-zhong, (1 School of Science, Shenyang Institute of Technology, Shengyang Liaoning 110168 China;2 Department of Mathematics, Beijing Institute of Technology, Beijing 100081 China)
Abstract:A two-step method is proposed to estimate the parameters in the Multi-binary response probit model.The √n-consistent estimation can be derived from marginal maximum likelihood in the first step, and then an asymptotic efficient estimation can be derived by one-step iteration in the second step.The simulation method is employed to estimate the Fisher information matrix in the second step. Due to the only one-step iteration, the random number of the simulation integral can be increased to diminish the error of simulation.
Keywords:LDV models: Multi-binary response models  Probit models  Maximum likelihood estimation  Asymptotic efficient estimation
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号