The Multiscale Parameter Estimation Methods for a Sort of Time Series  

作　　者：WEN Chenglin WEN Chenglin WANG Songwei WANG Songwei WEN Chuanbo WEN Chuanbo[4] CHEN Zhiguo[4] CHEN Zhiguo[4] 

机构地区：[1]Institute of Information and Control, Hangzhou Dianzi University, Hangzhou 310018, China [1]Institute of Information and Control, Hangzhou Dianzi University, Hangzhou 310018, China [2]Institute of Computer and Information Engineering, Henan University, Kaifeng 475001, China [2]Institute of Computer and Information Engineering, Henan University, Kaifeng 475001, China [3]College of Astronautics, Northwestern Polytechnical University, Xi＇an 710072, China [3]College of Astronautics, Northwestern Polytechnical University, Xi＇an 710072, China [4]Electric Engineering School, Shanghai Dianji University, Shanghai 200240, China [4]Electric Engineering School, Shanghai Dianji University, Shanghai 200240, China

出　　处：《Chinese Journal of Electronics》2009年第4期660-664,共5页电子学报（英文版）

基　　金：This work is mainly supported by the National Natural Science Foundation of China （No.60434020, No.60572051, No.60801048）, Tsinghua National Laboratory for Information Science and Technology （TNList） Cross- discipline Foundation, Zhejiang Province NSFC China （No.R106745, No.Y1080422）, International Cooperation Item of Zhejiang Province （No.2006C24G2040012）.

摘　　要：There are a lot of time series in many fields, especially, the long memory time series, and one of main tasks to research them is how to estimate corresponding se- ries parameters. There exist some current methods, such as the Traditional maximum likelihood estimation （TMLE） and Least square estimation （LSE） etc., but the huge com- putation burden is always a bottleneck to utilize them broadly in many applications. To overcome this difficulty, two new parameter estimation method, named identically Multiscale maximum likelihood estimation （MMLE）, are proposed by combining Discrete wavelet transform （DWT） and Discrete wavelet package transform （DWPT） with the TMLE in this paper, respectively. These primary ideas are all as follows, firstly, applying the DWT/DWPT to these time series, which possess of some good properties, such as orthogonal decomposition and decorrelation; see- ondly, analyzing the time series in a multiscale domain, and studying their statistical properties in different scale, such as meant variance and covariance. These new algorithms can effectively decrease computation complexity and ob- tain satisfying estimation precision illustrated by the data analysis and computer simulation.

关 键 词：Time Series Long memory process Discrete wavelet transform （DWT） Traditional maximum likelihood estimation （TMLE） Multiscale maximum like- lihood estimation （NIMLE） Computation complexity. 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] F830.9[自动化与计算机技术—计算机科学与技术]