Prior sensitivity analysis for Bayesian Markov chain Monte Carlo output

贝叶斯马尔可夫链蒙特卡罗输出的先验灵敏度分析

• 批准号: DP180102538
• 负责人: A/Prof Liana Jacobi
• 金额: $ 21.73万
• 依托单位: The University of Melbourne
• 依托单位国家: 澳大利亚
• 项目类别: Discovery Projects
• 财政年份: 2018
• 资助国家: 澳大利亚
• 起止时间: 2018-01-01 至 2023-12-31
• 项目状态: 已结题
• 来源: https://dataportal.arc.gov.au/RGS/Web/Grants/DP180102538
• 关键词: Prior; sensitivity; analysis; Bayesian; Markov

This project aims to develop the first set of techniques to implement an automated output sensitivity analysis for Markov Chain Monte Carlo (MCMC) estimation methods. Computationally intense Bayesian MCMC provide a powerful alternative to classical methods for the estimation of economic models. An obstacle to their wider application is that researchers need to specify prior beliefs about model parameters that will affect the results. The expected outcomes will enable researchers to undertake a routine assessment of the sensitivity of the results to prior inputs.

该项目旨在开发第一套技术，以实现马尔可夫链蒙特卡罗（MCMC）估计方法的自动输出灵敏度分析。计算密集型贝叶斯MCMC提供了一个强大的替代经典方法的经济模型的估计。其更广泛应用的一个障碍是，研究人员需要指定有关影响结果的模型参数的先验信念。预期结果将使研究人员能够对结果对先前输入的敏感性进行常规评估。