Premium
LEVINSON‐TYPE RECURSIVE ALGORITHMS FOR LEAST‐SQUARES AUTOREGRESSION
Author(s) -
Huang Dawei
Publication year - 1990
Publication title -
journal of time series analysis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.576
H-Index - 54
eISSN - 1467-9892
pISSN - 0143-9782
DOI - 10.1111/j.1467-9892.1990.tb00059.x
Subject(s) - autoregressive model , mathematics , algorithm , least squares function approximation , set (abstract data type) , explosive material , type (biology) , statistics , computer science , ecology , chemistry , organic chemistry , estimator , biology , programming language
. A set of formulae for calculating the least‐squares autoregressive coefficients is given. It can be used for stable, unstable and explosive models. The calculations needed by this algorithm are less than those of the Burg and Marple algorithms. The method used to deduce these formulae has general significance. For example, it is also used to improve the Marple algorithm. Simulation shows that the estimates given by this algorithm are better than those given by the Levinson and Burg recursions.