【摘要】 对多个相依统计模型的研究,现有成果主要集中在相依线性回归系统.本文则首次提出多个相依非线性回归系统中的附加信息Bayes拟似然,给出误差相关信息和先验信息在拟似然中的迭加方法,在较弱的条件下得到附加信息Bayes拟似然的一些性质,在Bayes风险准则下。讨论了其估计函数和参数估计的最优性,证明了附加信息Bayes拟似然的渐近 Bayes风险随着相依信息的增力。而逐步减少.更多还原

【Abstract】 Abstract It is well known that the recent theory on seemingly unrelated statistical sys-tems focuses mainly on linear regression systems. In this paper, we originate a Bayes quasi-likelihood with auxiliary information in some seemingly unrelated nonlinear re-gression systems. The method for iteratively adding related error information and prior information to an estimator is obtained. The optimal properties of its score function and parameter estimation based on the Bayes risk are discussed. It is shown that the asymptotic Bayes risk of Bayes quasi-likelihood estimation with auxiliary information decreases gradually with the increase of auxiliary information.更多还原