Title :
Further comments on nonstationarity identification problems for autoregressive models
Author_Institution :
Dept. of Stat., Georgia Univ., Atlanta, GA, USA
Abstract :
Nonstationarity for autoregressive models is identified by giving the necessary and sufficient conditions for autoregressive models to be oscillatory. The criteria previously proposed to identify autoregressive models to be asymptotically stationary or nonstationary are improved and extended to general multidimensional autoregressive models
Keywords :
identification; statistical analysis; autoregressive models; necessary conditions; nonstationarity identification; sufficient conditions; Artificial intelligence; Biological system modeling; Chemistry; Control system synthesis; Explosives; Multidimensional systems; Polynomials; Statistics; Sufficient conditions; System identification;
Conference_Titel :
Decision and Control, 1990., Proceedings of the 29th IEEE Conference on
Conference_Location :
Honolulu, HI
DOI :
10.1109/CDC.1990.203381
