双函数系数ARCH-M模型的正交加权经验似然检验

Orthogonal Weighted Empirical Likelihood Test of the Double-Functional Coeffcient ARCH-M Model

  • 摘要: 考虑一类双函数系数 ARCH-M 模型的检验问题. 通过构造正交加权的辅助随机向量, 提出了一种基于正交加权的经验似然检验方法. 在一些正则条件下, 理论证明了所构造的经验对数似然比检验统计量渐近服从卡方分布, 进而得到了一定置信水平的拒绝域, 最后通过数值模拟讨论了其检验功效.

     

    Abstract: The problem of testing a class of double-functional coeffcient ARCH-M models is considered in this paper. An empirical likelihood test based on orthogonal weighting is proposed by constructing an auxiliary random vector with orthogonal weighting. Under some regular conditions, it is theoretically proved that the constructed empirical log-likelihood ratio test statistic asymptotically obeys the chi-square distribution, and the rejection domain with a certain confidence level is obtained, and finally the test effcacy is discussed by numerical simulation.

     

/

返回文章
返回