Estimation based on one step ahead prediction versus estimation based on multi-step ahead prediction

15 December 1981

journal article
research article
Published by Taylor & Francis in Stochastics

Vol. 6 (1) , 43-55
https://doi.org/10.1080/17442508108833190

Abstract

In this paper we consider a strictly stationary time series generated by a nonlinear autoregression. We are concerned with the estimation of the parameter θ₀ which specifies the autoregression Two estimators are considered, namely. θ_n obtained by minimising the sum of squarcs of the sample prediction emets of a one step ahead predictor and θ_n obtained by minimising the sum of squares of the sample prediction errors of a multi-step ahead predictor. It is shown that θ_n is a better estimator of θ₀ than θ_n.

Keywords

This publication has 7 references indexed in Scilit:

Stationarity and invertibility of simple bilinear models
Stochastic Processes and their Applications, 1982
An Optimality Property of the Least-Squares Estimate of the Parameter of the Spectrum of a Purely Nondeterministic Time Series
The Annals of Statistics, 1980
Nonlinear autoregressive processes
Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences, 1978
Random Processes
Published by Springer Nature ,1974
Estimation and information in stationary time series
Arkiv för Matematik, 1953
On the Statistical Treatment of Linear Stochastic Difference Equations
Econometrica, 1943
Some Theorems on Distribution Functions
Journal of the London Mathematical Society, 1936