Title :
On the consistency of identification by dynamic factor models
Author :
Heij, C. ; Scherrer, W.
Author_Institution :
Tinbergen Inst., Erasmus Univ., Rotterdam, Netherlands
Abstract :
In this paper we present a new type of dynamic factor model. Here the observed process is decomposed into a factor part and a noise part. The factor part contains less freedom than the original process, in the sense that it satisfies linear system restrictions. Stated otherwise, the observed stochastic process is modelled by a deterministic behaviour. We pay special attention to consistency properties for this type of model
Keywords :
identification; least squares approximations; linear systems; stochastic processes; consistency; dynamic factor models; identification; least squares; linear system; stochastic process; Econometrics; Economic forecasting; Least squares methods; Linear systems; Operations research; Predictive models; Psychometric testing; Statistical analysis; Stochastic processes; System identification;
Conference_Titel :
Decision and Control, 1994., Proceedings of the 33rd IEEE Conference on
Conference_Location :
Lake Buena Vista, FL
Print_ISBN :
0-7803-1968-0
DOI :
10.1109/CDC.1994.411360
