ZENG Huifang,XIONG Peiyin.Bayesian analysis of the quantile AR-ARCH models based on MCMC algorithm[J].Journal of Yanbian University,2014,40(02):138-141.
基于MCMC的分位回归AR-ARCH模型的贝叶斯分析
- Title:
- Bayesian analysis of the quantile AR-ARCH models based on MCMC algorithm
- Keywords:
- Bayesian; quantile; AR-ARCH models; simulation
- 分类号:
- F224.9; O212
- 文献标志码:
- A
- 摘要:
- 针对时间序列分布特征的高峰厚尾性,提出了一类分位回归ARCH模型.在贝叶斯理论框架下,通过选择适当的先验分布,并基于非对称Laplace分布构建模型的似然函数,实现了模型的贝叶斯推断.仿真试验和分析表明,该分位回归ARCH模型可全面刻画时间序列的非对称性和高峰厚尾性.
- Abstract:
- Since many time series with asymmetric and heavier tails, we adapt the quantile regression ideas to the ARCH models. In the framework of Bayesian theory, we employ the proper prior, the likelihood function based on the asymmetric Laplace distribution was employed irrespective of the original distribution of the data, and derive the posterior distribution of the model parameters. The simulation result shows that the quantile ARCH models are effective to capture the diversity of time series distribution.
参考文献/References:
[1] Koenker R, Zhao Q. L-estimation for linear heteroscedastic models[J]. Journal of Nonparametric Statistics, 1994,3:223-235.
[2] Koenker R, Zhao Q. Conditional quantile estimation and inference for ARCH models[J]. Econometric Theory, 1996,12:793-813.
[3] Xiao Z, Koenker R. Conditional quantile estimation for generalized autoregressive conditional heteroscedasticity models[J]. Journal of the American Statistical Association, 2009,104(488):1696-1712.
[4] Chen C W S, Gerlach R, Wei D C M. Bayesian causal effects in quantiles: accounting for heteroscedasticity[J]. Computational Statistics & Data Analysis, 2009,53:1993-2007.
[5] 王新宇,宋学锋.间接TARCH-CAViaR模型及其MCMC参数估计与应用[J].系统工程理论与实践,2008(9):46-51.
[6] 曾惠芳,朱慧明,李素芳.基于MH算法的贝叶斯分位自回归模型[J].湖南大学学报:自然科学版,2010,37(2):88-92.
[7] 朱慧明,王彦红,曾惠芳.基于逆跳MCMC的贝叶斯分位自回归模型研究[J].统计与信息论坛,2010,25(1):9-13.
备注/Memo
收稿日期: 2014-03-11 作者简介: 曾惠芳(1981—),女,博士,讲师,研究方向为贝叶斯计量经济、极值理论.基金项目: 国家自然科学基金资助项目(41301421); 教育部人文社会科学研究项目(13YJC790203)