Threshold bounds in SVD and a new iterative algorithm for order selection in AR models
- 1 May 1991
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Signal Processing
- Vol. 39 (5) , 1218-1221
- https://doi.org/10.1109/78.80960
Abstract
The problem of order determination of AR (autoregressive) models using singular value decomposition (SVD) is reexamined from a statistical point of view. Thresholds for distinguishing between significant and nonsignificant singular values are derived, and a novel iterative algorithm for order selection in AR models is presented. Simulation results show the technique to be very effective when a small number of samples is availableKeywords
This publication has 7 references indexed in Scilit:
- Accuracy of frequency estimation and its relation to prediction filter orderPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- Applications of singular value decomposition to system modeling in signal processingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- Statistical analysis of effective singular values in matrix rank determinationIEEE Transactions on Acoustics, Speech, and Signal Processing, 1988
- Order selection for AR models by predictive least squaresIEEE Transactions on Acoustics, Speech, and Signal Processing, 1988
- Singular value decomposition and improved frequency estimation using linear predictionIEEE Transactions on Acoustics, Speech, and Signal Processing, 1982
- Modeling by shortest data descriptionAutomatica, 1978
- A new look at the statistical model identificationIEEE Transactions on Automatic Control, 1974