NGINAR(1)模型参数的拟似然估计  

Quasi-Likelihood Estimation of Parameters in NGINAR( 1) Model

作　　者：贺天宇[1] 王金山[1] 

机构地区：[1]陆军军官学院,合肥230031

出　　处：《重庆理工大学学报（自然科学）》2015年第2期141-146,共6页Journal of Chongqing University of Technology：Natural Science

基　　金：军内基金资助项目(jnky2012015;2014XYJJ-032)

摘　　要：利用拟似然法对Ristic提出的几何分布的整值自回归过程(NGINAR(1))的参数估计问题进行研究,得到了参数的修正拟似然估计因子及其极限分布,并做了数值实验,将拟似然估计与Yule-Walker估计及CLS估计进行比较。结果表明:拟似然估计在一定程度上优于Yule-Walker估计及CLS估计。We used quasi-likelihood method to estimate the parameter of NGINAR （ 1 ） process, and obtained the modified-quasi-likelihood estimator of the model parameters and the asymptotic distribu- tion of this estimator. We compared the modified-quasi-likelihood estimator with the Yule-Walker estimator and the conditional least squares estimator via simulation. The results show that the modified- quasi-likelihood estimator is better than Yule-Walker estimator and the conditional least squares estimator.

关 键 词：整值时间序列 负二项稀疏算子 拟似然估计 渐近分布 

分 类 号：O212.7[理学—概率论与数理统计]