ASYMPTOTIC OPTIMAL INFERENCE FOR A CLASS OF NONLINEAR TIME-SERIES MODELS

Authors
HWANG SY BASAWA IV
Citation
Sy. Hwang et Iv. Basawa, ASYMPTOTIC OPTIMAL INFERENCE FOR A CLASS OF NONLINEAR TIME-SERIES MODELS, Stochastic processes and their applications, 46(1), 1993, pp. 91-113
Citations number
19
Categorie Soggetti
Statistic & Probability","Statistic & Probability
Journal title
Stochastic processes and their applications → ACNP
ISSN journal
03044149
Volume
46
Issue
1
Year of publication
1993
Pages
91 - 113
Database
ISI
SICI code
0304-4149(1993)46:1<91:AOIFAC>2.0.ZU;2-S
Abstract
The local asymptotic normality (LAN) of the log-likelihood ratio for a class of Markovian nonlinear time series models is established using the approach of quadratic mean differentiability. The error process in the models considered is not necessarily Gaussian. As a consequence o f the LAN property, asymptotically optimal estimators of the model par ameters are derived. Also, asymptotically efficient tests for linearit y are constructed. Several examples are discussed as special cases.