Fifth International Workshop on Objective Bayesian Methodology

第五届国际客观贝叶斯方法研讨会将于2005年6月4-8日在密苏里州布兰森举行。客观贝叶斯方法在很大程度上是面向开发可自动使用的先验分布，即除了选择特定的概率模型来描述数据之外，不需要主观输入。有三个截然不同的统计领域出现了这种发展：参数估计、模型选择和预测，客观贝叶斯方法在今天越来越重要，因为贝叶斯分析在非专业人士中的应用迅速增长，他们中的大多数人寻求自动或客观的贝叶斯程序。本研讨会将强调预测、实际应用，如时空模型、多重比较和拟合优度。贝叶斯方法通常严重依赖于先验分布。尽管先验分布是贝叶斯范式的重要组成部分，众所周知，即使在没有任何相关先验信息的情况下，贝叶斯方法也可以通过采用适当的客观（非信息）贝叶斯分析来对许多具有挑战性的问题产生可靠的推断。其主要目标是促进客观贝叶斯方法论的最新研究进展的交流，为新的研究人员提供机会，并建立新的合作，将努力引导到悬而未决的问题，并开辟新的调查方向。这次会议的一个重要结果将是进一步明确客观先验方法论作为统计研究的一个既定领域。统计学家传统上接受两种主要方法之一，经典或“频率论”方法和贝叶斯方法。近年来，计算技术的进步和统计理论的突破使得贝叶斯方法越来越有吸引力，特别是对于气象学、流行病学、自然资源等领域中具有许多变量的困难问题。随着对贝叶斯统计的日益重视，弥合两个统计思想流派之间的差异变得越来越重要。 目的贝叶斯学派致力于解决这些差异。 计划举办的讲习班将汇集80至100名与会者，包括来自世界各地的客观贝叶斯统计领域的领导人，以及年轻的研究人员和学生，在一个特别有利于合作和交流思想的环境中。

The Fifth International Workshop on Objective Bayesian Methodology willbe held in Branson, Missouri, June 4-8, 2005.Objective Bayesian methodology is, for the most part, oriented towardsthe development of prior distributions that can be used automatically,i.e. that do not require subjective input other than the specificprobabilistic model chosen to describe the data. There are three quitedistinct statistical domains in which this development has taken place:parametric estimation, model selection, and prediction,Objective Bayesian methodology is of increasing importance today sinceapplication of Bayesian analysis is rapidly growing among nonspecialists,most of whom seek automatic or objective Bayesian procedures.This workshop will emphasize prediction, practical applicationssuch as spatial-temporal models, multiple comparisons and goodness-of-fit.Bayesian methods usually are heavily depend on prior distributions.Although a prior distribution is an important component in the Bayesianparadigm, it is known that even in the absence of any relevant priorinformation, the Bayes method can produce reliable inference to manychallenging problems by employing appropriate objective (noninformative)Bayesian analysis. The principal objectives are to facilitate theexchange of recent research developments in objective Bayesianmethodology, to provide opportunities for new researchers, and toestablish new collaborations that will channel efforts into pendingproblems and open new directions for investigation. An importantconsequence of this meeting will be to further crystallize objective priormethodology as an established area for statistical research.Statisticians have traditionally embraced one of two main approaches,classical or "frequentist" methods and Bayesian methods. In recent years,advances in computing technology and breakthroughs in statistical theoryhave made Beyesian methods increasingly attractive, especially for difficult problems with many variables in fields such as meteorology, epidemiology, natural resources, etc. With this increased emphasis on Bayesian statistics, it has become increasingly important to bridge the differences between the two schools of statistical thinking. Objective Bayesians are dedicated to resolving these differences. The planned workshop will bring together 80-100 participants including leaders in the field of objective Bayesian statistics from around the world together with young researchers and students in a setting especially conducive to cooperation and exchange of ideas.