摘要
目的：提出一种客观的统计检验方法来判断时间序列的平稳性。方法：根据Ｆｕｌｅｒ提出的τμ和ττ分布，采用回归分析的方法解决判断时间序列的平稳性。结果：统计检验的方法与传统的图示方法的结果是一致的。避免了主观性及过度差分。结论：统计方法目前仅适用于没有季节性波动的ＡＲＩＭＡ模型。
Objective:This paper gives a statistical method of how to test if a time series is stationary.Methods:Based on the Fuller' τ μ and τ τ distribution,we can test if a time series is stationary by regression analysis.Results:The statistical test method is consistent with the traditional method,and can overcome subjective shortcoming of traditional method and overdifferencing.Conclusions:The statistical test method is only suit for nonseasonal ARIMA model.It is good choice to judge stationarity of a time series with using both the statistical test method and traditional methods.
出处
《中国卫生统计》
CSCD
北大核心
1998年第3期12-14,共3页
Chinese Journal of Health Statistics
