Order estimation by accumulated prediction errors
- 1 January 1986
- journal article
- Published by Cambridge University Press (CUP) in Journal of Applied Probability
- Vol. 23 (A) , 55-61
- https://doi.org/10.2307/3214342
Abstract
This paper presents a new criterion based on prediction error which allows the estimation of the number of parameters as well as structures in statistical models. The criterion is valid for short and long samples alike. Unlike Akaike's earlier criterion, also based on prediction error, the criterion proposed here appears to produce consistent error estimates in ARMA processes.Keywords
This publication has 13 references indexed in Scilit:
- Universal coding, information, prediction, and estimationIEEE Transactions on Information Theory, 1984
- Properties of Predictors for Autoregressive Time SeriesJournal of the American Statistical Association, 1981
- The Estimation of the Order of an ARMA ProcessThe Annals of Statistics, 1980
- Asymptotically Efficient Selection of the Order of the Model for Estimating Parameters of a Linear ProcessThe Annals of Statistics, 1980
- A Predictive Approach to Model SelectionJournal of the American Statistical Association, 1979
- The Determination of the Order of an AutoregressionJournal of the Royal Statistical Society Series B: Statistical Methodology, 1979
- Modeling by shortest data descriptionAutomatica, 1978
- Asymptotic normality of prediction error estimators for approximate system modelsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1978
- A new look at the statistical model identificationIEEE Transactions on Automatic Control, 1974
- The prediction error of stationary Gaussian time series of unknown covarianceIEEE Transactions on Information Theory, 1965