非平稳和长记忆时间序列主频率估计方法研究被引量：1

Study on Estimation of Dominating Frequency of Non-Stationary and Long Memory Time Series

下载PDF

导出

摘要许多经济时间序列表现出非平稳性和长记忆性,对这样的序列进行谱分析并估计其主频率,采用传统的谱分析方法已不适用.基于平稳和非平稳分数阶差分自回归滑动平均模型,研究了具有非平稳性和长记忆性序列的主频率的估计方法,同时提出了包括模型定阶在内的基于极小化残差平方和的近似极大似然参数估计算法.通过对上海和深圳证券交易所综合指数收益序列的分析,说明了此方法的可行性和有效性. Many economic time series are nonstationary and have long memory. Traditional spectral analytical methods are not fit for the spectral analysis or the estimation of the dominating frequency of such series. The estimation method of the dominating frequency of such series is studied on the basis of the invertible short and long memory autoregressive integrated moving average model. An algorithm including the determination of the model orders on approximate maximum likelihood parameters estimation method of minimizing the sum of the squared residuals is put forward. The method suggested is proved to be effective and reasonable by the analysis of the income series of aggregate index of Shanghai and Shenzhen Stock Markets.

作者徐梅张世英樊智

机构地区天津大学管理学院

出处《天津大学学报（自然科学与工程技术版）》 EI CAS CSCD 北大核心 2003年第4期507-511,共5页 Journal of Tianjin University：Science and Technology

基金国家自然科学基金资助项目(70171001).

关键词经济时间序列非平稳序列长记忆序列主频率估计方法谱分析差分自回归滑动平均模型近似极大似然估计 spectral analysis dominating frequency non-stationary time series long memory time series

分类号 F224.7 [经济管理—国民经济]

引文网络
相关文献

参考文献8

1Priestley M B, Spectral Analysis and Time Series [ M ]. London: Academic Press, 1981.
2Jan Beran, Sucharita Ghosh. Estimation of dominating frequency for stationary and nonstationary fractional autogressive models[J]. Journal of Time Series Analysis, 2000,21(5) : 517--533.
3Granger C W J, Joyeux R. An introduction to long-range time series models and fractional differencing [ J ]. Journal of Time Series Analysis, 1980, 1 ( 1 ) : 15--30.
4Brockwell P J, Davis R A. Time Series : Theory and Methods[ M ]. New York : Springer-Verlag, 1991.
5Granger C W J, Zhuan Xin Ding. Varieties of long memory models[J]. Journal of Econometrics, 1996, 73 : 61--78.
6Baillie R T. Long memory process and fractional integration in econometrics [ J ]. Journal of Econometrics, 1996, 73 :5--59.
7Jan Beran. Maximum likelihood estimation of differencing parameter for invertible short and long memory autoregressive integrated moving average models [ J ]. J R Statist Sco B, 1995,57(4) : 659----672.
8Box G E, Jenkins G M. Time Series Analysis: Forecasting and Control[ M]. San Francisco: Holden Day, 1976.