基于异方差检验的水文过程隐含周期分析模型及其应用--Ⅰ：模型  

Crytic period analysis model of hydrological process based on herteroskedasticity test and its application Ⅰ. Model

作　　者：王红瑞[1] 林欣[2] 钱龙霞[1] 张淑梅[2] 

机构地区：[1]北京师范大学水科学研究院,水沙科学教育部重点实验室,北京100875 [2]北京师范大学数学科学学院,北京100875

出　　处：《水利学报》2008年第11期1183-1189,共7页Journal of Hydraulic Engineering

基　　金：国家科技支撑计划项目(2006BAB04A08)

摘　　要：本文在讨论水文时序平稳性和异方差性的基础上,提出了检测水文时序隐含周期的模型,即对水文时序设计对数变换序列,提高序列的平稳性;建立相应的自回归模型降低残差的异方差性;通过对所得平稳序列进行周期图分析确定可能的主周期分量。针对水文时序的特性提出包括具有确定频率的正弦项存在性检验、具有确定整数周期的非正弦周期项存在性检验和非确定频率的隐含周期性检验(Fisher检验)三种检验方式,可择其中一种确定主周期的显著水平。采用该模型可以解决利用自相关函数判断序列周期性所不能解决的问题,可进一步挖掘数据隐含的、深层次信息。最后,本文还编制了水文时序隐含周期的计算流程。在理论上,该模型不仅解决了隐含周期的检测问题,对于有确定周期的序列亦适用。A model for finding the cryptic period of hydrological time series is proposed. The functions of the model include to apply logarithm transformation for a corresponding auto-regression model for reducing the making the sequence more stationary, to construct herteroskedasticity of residual, to determine the component of the prime period by periodogram analysis. According to the characteristics of hydrological time series, three kinds of tests for determining the significance level of the prime period are suggested including the sine term existence test with deterministic frequency, non-sine period term existence test with deterministic integral period and cryptic periodicity test with nondeterministic frequency. The model can be used to deal with the problems that are beyond the autocorrelation function and to mine the cryptic and deep information. Moreover, the flow chart of proposed calculating the cryptic period is established. The model not only can be used to test the cryptic period but also applicable to the sequences which have deterministic periods.

关 键 词：水文时序 自回归模型 异方差 隐含周期 平稳性 

分 类 号：P333[天文地球—水文科学]