Evaluation of Bayesian source estimation methods with Prairie Grass observations and Gaussian plume model: A comparison of likelihood functions and distance measures

详细信息    查看全文

• 作者：Yan Wang^a ; Hong Huang^a ; ^{hhong@tsinghua.edu.cn} ; Lida Huang^a ; Branko Ristic^b
• 关键词：Bayesian inference ; Source estimation ; Approximate Bayesian computation ; Markov chain Monte Carlo ; Sequential Monte Carlo ; Prairie Grass field experiment
• 刊名：Atmospheric Environment
• 出版年：2017
• 出版时间：March 2017
• 年：2017
• 卷：152
• 期：Complete
• 页码：519-530
• 全文大小：2369 K
• 卷排序：152

文摘

Slice sampling is introduced to solve Bayesian source estimation problems. Approximate Bayesian Computation method is applied to source estimation problems. Five likelihood functions are evaluated using field experiment data sets. Six distance measures are evaluated using field experiment data sets. Nemenyi test is used to evaluate all the methods over multiple data sets.