Premium
A NEW WAY TO ESTIMATE ORDERS IN TIME SERIES
Author(s) -
Zhang HuMing,
Wang Ping
Publication year - 1994
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.1467-9892.1994.tb00210.x
Subject(s) - akaike information criterion , identifiability , bayesian information criterion , mathematics , information criteria , series (stratigraphy) , statistics , bayes' theorem , least squares function approximation , order (exchange) , estimation , econometrics , algorithm , mathematical optimization , model selection , bayesian probability , paleontology , management , finance , estimator , economics , biology
. In this paper we propose the order determination quantity (ODQ) as a new way to solve order estimation problems in time series analysis. We estimate orders according to ODQ > 0 or ODQ < 0 instead of by minimizing. Theoretical analysis and simulation have shown that the ODQ has higher identifiability for unknown true orders, provides clear separation points and requires less computational effort than the existing order estimation criteria such as Akaike's information criterion (AIC), Bayes information criterion (BIC), φ and predictive least squares (PLS).