摘要
为了解决候选模型较多,无法一一比较其准则值的问题,提出基于Gibbs样本生成器(Gibbs sampler)的广义自回归条件异方差(GARCH)模型的选择方法.模拟实验结果表明:该模型选择方法可以高效、准确地从大量的候选模型中选择出准则值最小的模型.
In order to solve the problem of more candidate models that we can′t compare the criterion values one by one,we put forward a selecting method of the GARCH（generalized auto-regressive conditional heteroskedasticity） model based on Gibbs sampler.The method estabish a connection between criterion values of candidate models and probabilities of candidate models.When the number of the models generated becomes large enough,the model with the lowest criterion value will tend to appear early and frequently.The result shows that we can choose the model with the lowest criterion value,accurately and efficiently,from the candidate models.
出处
《华侨大学学报（自然科学版）》
CAS
北大核心
2011年第4期443-446,共4页
Journal of Huaqiao University(Natural Science)
基金
福建省自然科学基金资助项目(2009J01312)
