State space models and estimation of missing observations in time series

Loading...
Thumbnail Image
Date
2012-02-24
Authors
Biwott, Daniel Kiprotich
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In this project we have considered a non-linear time series model, which encompasses several standard non-linear models for time series as special cases. It also offers two methods for estimating missing observations, one using prediction and fixed point smoothing algorithm and the other using optimal estimating equation theory. Recursive estimation of missing observations in an Autoregressive Conditionally Heteroscedastic (ARCH) model and the estimation of missing observations in a linear time series model are shown to be special cases. For the case of prediction and fixed point smoothing algorithm, we have generalised the formula developed by Abraham and Thavaneswaran (1991) for estimating missing observations to a case when there are more than two missing observations. Simulation studies have been carried out on AR 91) data to illustrate the application of the formula.
Description
Department of Mathematics, 94 p. The QA 280.B5 2000
Keywords
Time-series analysis, State-space methods
Citation