一种基于AEPD的贝叶斯参数估计方法
A Technique of Bayesian Parameter Estimation Based on AEPD
摘要:
本文基于非对称指数幂分布(AEPD)，采用MCMC中的M-H算法进行贝叶斯参数估计，针对后验密度的复杂性，采取随机游走链的建议分布为中心对称分布。通过数值模拟，在M个大样本下进行N次迭代，模拟结果显示，该算法对具有非对称指数幂分布的贝叶斯参数估计具有实用性。
Abstract:
In this paper, based on the asymmetric exponential power distribution (AEPD), the M-H algorithm in MCMC is used to estimate the Bayesian parameters. For the complexity of the posterior density, the recommended distribution of the random walk chain is the center symmetric distribution. Through numerical simulation, N iterations are performed under M large samples. The simulation results show that the algorithm is practical for Bayesian parameter estimation with asymmetric exponential power distribution.
参考文献
|
[1]
|
Zhu, D. and Zinde-Walsh, V. (2009) Properties and Estimation of Asymmetric Exponential Power Distribution. Journal of Econometrics, 14, 86-99.
[Google Scholar] [CrossRef]
|
|
[2]
|
Chib, S. and Greenberg, E. (1995) Understanding the Metropolis-Hastings Algorithm. The American Statistician, 49, 327-335.
[Google Scholar] [CrossRef]
|
|
[3]
|
Altaleb, A. and Chauveau, D. (2002) Bayesian Analysis of the Logit Model and Comparison of Two Metropolis-Hastings Strategies. Computational Statistics & Data Analysis, 39, 137-152.
[Google Scholar] [CrossRef]
|
|
[4]
|
Geweke, J. and Tanizaki, H. (2003) Note on the Sampling Distribution for the Metropolis-Hastings Algorithm. Communications in Stats, 32, 775-789.
[Google Scholar] [CrossRef]
|
|
[5]
|
Minh, D.D.L. and Minh, D.L. (2015) Understanding the Hastings Algorithm. Communications in Statistics, B. Simulation and Computation.
[Google Scholar] [CrossRef]
|