摘要
为解决时间序列因观测时间较短而难以确定变点的问题,通过分析独立同分布序列面板数据,对其均值的渐变变点进行检验.假设面板数据中每条时间序列的均值在同一时刻发生变化,提出Ratio检验统计量,并分别在原假设与备择假设下给出检验统计量的极限性质.数值模拟结果表明,面板数据的检验效果优于单条时间序列,实例分析也验证了该方法的有效性与实用性.
In order to solve the problem that it is difficult to determine the change point of time series due to the short observation time,the gradual changes of means in panel data are tested by analyzing the panel data of independent identically distributed series.Assuming that the mean of each series in the panel data changes at the same time,the Ratio test statistics are proposed,and the limit properties are obtained under the null and alternative hypothesis respectively.The results of numerical simulation show that the test effect on panel data is better than that on a single series,and the example analysis also verifies the effectiveness and practicability of this method.
作者
刘鑫
赵文芝
李建辉
LIU Xin;ZHAO Wenzhi;LI Jianhui(School of Computer Science,Xijing University,Xi'an 710123,China;School of Science,Xi’an Polytechnic University,Xi'an 710048,China)
出处
《扬州大学学报（自然科学版）》
CAS
北大核心
2023年第1期20-24,共5页
Journal of Yangzhou University：Natural Science Edition
基金
陕西省自然科学基础研究资助项目(2021JQ-867)
陕西省教育厅科研计划资助项目(22JK0591)
西京学院科研基金资助项目(XJ180209)。
