吉林大学学报(理学版) ›› 2024, Vol. 62 ›› Issue (4): 866-877.

• • 上一篇    下一篇

k阶Poisson相依驱动随机系数混合算子整数值自回归模型

刘秀芳1, 张晓磊2, 王德辉3   

  1. 1. 太原理工大学 数学学院, 太原 030024; 2. 吉林大学 数学学院, 长春 130012;3. 辽宁大学 数学与统计学院, 沈阳 110036
  • 收稿日期:2024-01-08 出版日期:2024-07-26 发布日期:2024-07-26
  • 通讯作者: 张晓磊 E-mail: zhangxl_jlu@163.com

k-Order Poisson Dependent-Driven Random Coefficient Mixed Thinning Integer-Valued Autoregressive Model

LIU Xiufang1, ZHANG Xiaolei2, WANG Dehui3   

  1. 1. College of Mathematics, Taiyuan University of Technology, Taiyuan 030024, China;
    2. College of Mathematics, Jilin University, Changchun 130012, China; 3. School of Mathematics and Statistics, Liaoning University, Shenyang 110036, China
  • Received:2024-01-08 Online:2024-07-26 Published:2024-07-26

摘要: 采用k阶Poisson相依驱动随机系数混合算子整数值自回归模型分析具有变量字符元素计数特征的数据, 给出该模型的统计性质和参数的条件极大似然估计, 并证明估计量的渐近正态性. 数值模拟结果表明, 随着样本容量的增加, 参数估计逐渐收敛于真实值. 实际数据分析结果表明了该模型的有效性.

关键词: k阶自回归模型, Po-DDRCMTINAR(k)模型, 混合稀疏算子, 条件极大似然估计

Abstract: By using k-order Poisson dependent-driven random coefficient mixed thinning integer-valued autoregressive model, we analyzed data with the counting of elements of variable character, gave  statistical properties of the model and conditional maximum likelihood estimation of parameters, and proved the asymptotic normality of the estimators. The numerical simulation results show that as the sample size increases, the parameter estimation gradually converges to the true value. The actual data analysis results  show  that the effectiveness of the proposed model.

Key words: k-order autoregressive model, Po-DDRCMTINAR(k) model, mixed thinning operator, conditional maximum likelihood estimation

中图分类号: 

  • O212.1