Mathematical Sciences: An Interdisciplinary Meeting on Recent Developments in the Theory and Application of Markov Chain Monte Carlo Numerical Models

数学科学：马尔可夫链蒙特卡罗数值模型理论与应用最新发展跨学科会议

• 批准号: 9629834
• 负责人: Arnold Zellner
• 金额: $ 0.96万
• 依托单位: University of Chicago
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1996
• 资助国家: 美国
• 起止时间: 1996-08-15 至 1997-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9629834&HistoricalAwards=false
• 关键词: Mathematical; Sciences; Interdisciplinary; Meeting; Recent

Zellner 9629834 the investigator and colleagues organize a workshop on Markov chain Monte Carlo (MCMC) techniques. These techniques, originally developed in mathematics and physics, have come to be of great importance in many sciences and fields of application. Their use allows researchers and applied workers to solve many theoretical and applied problems. While MCMC techniques have already been shown to be of great value, more work has to be done to improve them. For example, new methods are being developed to speed convergence of MCMC techniques, to check on the convergence and accuracy of MCMC output, and to provide computer systems and programs needed to implement MCMC techniques in a user-friendly, interactive environment. These are a subset of the topics, including new applications, that are discussed at the workshop. the investigator and colleagues organize a workshop on Markov chain Monte Carlo (MCMC) techniques. Monte Carlo techniques are ways to make probabilistic estimates of quantities. For example, the volumes of a complicated shape can be estimated by making random samples of the shape when embedded in a simpler shape. MCMC techniques have already been of great value in solving important problems in science, industry and government. In this connection, models are employed to explain past behavior, predict future outcomes and evaluate proposed policies. Deriving good solutions involves needed calculations that are best carried out by use of MCMC techniques in many instances, e.g. in connection with the analysis of global warming, financial asset-pricing, industry personnel decisions, on-line quality control and governmental policy problems. Better MCMC techniques, and more informed users of them, contribute to better solutions to these and other private and public scientific and policy problems.

Zellner 9629834研究人员及其同事组织了马尔可夫链蒙特卡洛（MCMC）技术的研讨会。 这些技术最初是在数学和物理学中开发的，在许多科学和应用领域都非常重要。 他们的使用使研究人员和应用工人可以解决许多理论和应用问题。 尽管已经证明MCMC技术具有很大的价值，但必须做更多的工作以改善它们。 例如，正在开发新方法，以加快MCMC技术的收敛性，检查MCMC输出的收敛性和准确性，并提供在用户友好的交互式环境中实现MCMC技术所需的计算机系统和程序。 这些是在研讨会上讨论的主题的子集，包括新应用程序。 研究人员和同事组织了马尔可夫链蒙特卡洛（MCMC）技术的研讨会。 蒙特卡洛技术是对数量进行概率估计的方法。 例如，复杂形状的体积可以通过在更简单的形状嵌入形状的随机样本中估算。 MCMC技术在解决科学，工业和政府中的重要问题方面已经具有很大的价值。 在这方面，使用模型来解释过去的行为，预测未来的结果并评估拟议的政策。 得出良好的解决方案涉及所需的计算，这些计算是在许多情况下使用MCMC技术最佳进行的，例如与分析全球变暖，金融资产定价，行业人员决策，在线质量控制和政府政策问题有关。 更好的MCMC技术，以及更多知识的用户，为这些以及其他私人和公共科学和政策问题提供了更好的解决方案。