Total Least Squares Method to Solve the Parameter Estimation of Autoregression Model
DOI:
https://doi.org/10.54097/fxdha146Keywords:
Autoregressive model, parameter estimation, total least squares, least squares.Abstract
Parameter estimation of autoregressive models is a fundamental problem in time series analysis. The total least squares estimation method can simultaneously take into account the errors of the coefficient matrix X and the observation vector Y, while correcting for both. In this paper, the total least squares is applied to the parameter estimation of the autoregressive AR model. The application example verifies the feasibility and effectiveness of the total least squares method used for the parameter estimation of the of the autoregressive model.
Downloads
References
Chuan Hu, Yi Chen, Parameters Estimation of Autoregression with Structured Total Least Squares, Journal of Geodesy and Geodynamics.2013, 33(1): 45-47.
Yibin Yao, Shuhua Huang, Jiajun Chen, A New Method of TLS to Solving the Autoregressive Model Parameter, Geomatics and Information Science of Wuhan University. 2014, 39(12): 1463-1466.
Markovsky I, Van Huffel S, Overview of Total Least-squares Methods. Signal Processing. 2007, 87(10): 2283-2302.
Yang G, Xie Z, Research on the algorithm of parameters estimation for linear regression on total least squares, Geomat Spat Inf. 2020, 43(1): 82–84.
HUANG Shengxiang, YIN Hui, JIANG Zheng, Deformation monitoring data processing. Wuhan: Wuhan University Press, 2010.
Downloads
Published
Issue
Section
License
Copyright (c) 2024 Highlights in Science, Engineering and Technology
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
