Premium
A note on Model Reference Adaptive System (MRAS) estimate with infinite variance
Author(s) -
Thavaneswaran A.,
Abraham B.
Publication year - 1994
Publication title -
statistica neerlandica
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.52
H-Index - 39
eISSN - 1467-9574
pISSN - 0039-0402
DOI - 10.1111/j.1467-9574.1994.tb01447.x
Subject(s) - mras , variance (accounting) , mathematics , autoregressive model , estimation , estimation theory , autoregressive–moving average model , reference model , variance components , statistics , control theory (sociology) , computer science , artificial intelligence , control (management) , physics , accounting , management , vector control , software engineering , quantum mechanics , voltage , economics , induction motor , business
This paper discusses the estimation of a parameter in an autoregressive model with infinite variance. A recursive estimation procedure based on minimizing the prediction errors is provided. It is also shown that the model reference adaptive system estimate for an AR (1) model given in Aase (1983) is a special case.