广义线性模型中带有测量误差的序贯压缩估计.pdfVIP

第45卷第6 期 • æ 绅筝在ß.大军牵息 Vol. 45 ,No. 6 JαJ刚AL OF UNIVEI芯ITYOF 比旧JCE AND TECHNOLC后Y OFCHINA Jun. 2 0 1 5 2015 年 6 月 Article ID: 0253-2778(2015)06-0449-11 Sequential shrinkage estimation in generalized linear models with measurement errors 1 1 1 LU Haiho ,2 , WANG Zhanfeng , WU Yaohua (1. De户t. of Statistics and Finance , Schωlof 品也nagement ， University of Science and Technology of αina ， Hefei 230026 ， αlna; 2. School of Mathematical Sciences , Xinjiang Normal University , Urumqi 830054 ， αina) Abstract: A sequential shrinkage estimation method was developed to determine a minimum sample size under which both of the variable selection and the parameter estimation with a pr~ specified accuracy were achieved for the generalized linear model with measurement errors. Asymptotic properties of the proposed sequential estimation method , such as the coverage probability of the sequential confidence set and the efficiency of the minimum sample size , were studied. Simulation studies were conducted and the results show that the proposed method can save a large number of samples compared to the traditional sequential sampling method. Finally a diabetes data set was used as an example. Key words: generalized linear model; sequential sampling; adaptive shrinkage estimate; stopping rule; confidence set CLC number:0212. 1 Document code:A doi: 10. 3969/j. issn. 0253-2778. 2015. 06. 004 2010 Mathemati臼 Subject Classification: Primary 62F12;