摘要
提出了基于MCMC方法来估计相关系数平稳序列模型的参数;给出基于贝叶斯分布的相关系数平稳序列模型参数的算法;在无信息先验分布下,模拟证明了用此方法估计相关系数平稳序列模型参数的优良效果。最后对实际的广西电网-月负荷数据,分别用基于相关系数平稳序列模型的MCMC方法和极大似然估计法以及基于经典的ARMA模型建模,结果表明采用MCMC方法得到的模型给出的预测是最好的。
MCMC method was proposed to estimate the parameters of the correlation coefficient stationary series. The algorithm of using Bayesian distribution to estimate the model parameters was given. Extensive simulation experiments have shown that the Bayesian estimation procedure under the non-informative prior distribution works well. The MCMC method and the maximum like hood estimation method on the correlation coefficient stationary series and model on the classical ARMA were applied on the real electric net load monthly data of Guangxi. The results show that the MCMC method provides the most precise prediction.
出处
《系统仿真学报》
EI
CAS
CSCD
北大核心
2008年第14期3648-3651,3655,共5页
Journal of System Simulation
关键词
相关系数平稳序列
MCMC模拟
贝叶斯估计
Gibbs抽样算法
电网负荷
correlation coefficient stationary series
MCMC simulation
Bayesian estimation
Gibbs sampling algorithms
electric net monthly load.
