| 基于MCMC的PLP未来强度的Bayesian预测分析 |
| |
| 引用本文： | 王燕萍,吕震宙,赵新攀.基于MCMC的PLP未来强度的Bayesian预测分析[J].航空计算技术,2010,40(2):1-4,27. |
| |
| 作者姓名： | 王燕萍 吕震宙 赵新攀 |
| |
| 作者单位： | 西北工业大学,航空学院,陕西,西安,710072 |
| |
| 基金项目： | 国家自然科学基金资助，民口863计划课题资助，新世纪优秀人才支持计划资助，航空基金资助 |
| |
| 摘 要： | 在无信息先验分布下，将Gibbs抽样与Metropolis-Hastings算法相结合的方法应用于幂律过程的未来强度的Bayesian预测。该预测方法能将时间截尾数据和失效截尾数据统一分析，并给出在未来某一时刻处强度函数的MCMC样本，利用该样本可以方便地获得关于未来某一时处刻强度函数及其函数的各种后验分析。一个经典工程数值算例说明了预测方法的可行性、合理性与有效性。
|
| 关 键 词： | 幂律过程 强度函数 Bayesian推断 Gibbs抽样 Metropolis—Hastings算法 预测分析 |
Bayesian Prediction Analysis of the Future Intensity of the Power Law Process Based on Markov Chain Monte Carlo |
| |
| WANG Yan-ping,LV Zhen-zhou,ZHAO Xin-pan.Bayesian Prediction Analysis of the Future Intensity of the Power Law Process Based on Markov Chain Monte Carlo[J].Aeronautical Computer Technique,2010,40(2):1-4,27. |
| |
| Authors: | WANG Yan-ping LV Zhen-zhou ZHAO Xin-pan |
| |
| Institution: | WANG Yan- ping, LV Zhen- zhou, ZHAO Xin- pan (School of Aeronautics ,Northwestern Polytechnical University ,X i＇an 710072, China) |
| |
| Abstract: | Under various reasonable noninformative priors, the hybrid of Gibbs sampling and Metropolis- Hastings algorithm is employed to prediction of Bayesian approach based the intensity of the power law process in the future time. The prediction could provide a unified analysis for both time and failure truncated data, and give Markov Chain Monte Carlo （ MCMC ） samples of the intensity in the future time. All kinds of posterior analysis about the intensity and functions of the intensity in the future time are obtained easily on these MCMC samples. Results from an engineering numerical example illustrate the feasibility, rationality and validity of the prediction. |
| |
| Keywords: | power law process intensity function bayesian inference gibbs sampling metropolis - hastings algo- rithm prediction analysis |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
