Identification and estimation of a first order bilinear time series

Afariebor, Roland (2001) Identification and estimation of a first order bilinear time series. Masters thesis, Memorial University of Newfoundland.

[English] PDF (Migrated (PDF/A Conversion) from original format: (application/pdf)) - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (7MB) [English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. (Original Version)

Abstract

In this practicum, we study the properties of a special case of the general bilinear model. The general bilinear model was proposed by Granger and Andersen(1978) and Subba Rao(1981) for studying non-linear time series. Simulation studies and real life data sets are used to evaluate the performance of the theoretical results we derived. The properties we study are the mean, covariance structures, third order moments and cumulants. We find a pattern in the third-order cumulant which is useful in identifying the order of the model. This work is an extension of the result of Oyet(2001). The model is used to make forecasts on three real time series data. -- Also considered are the mean and covariance structures of three other versions of the bilinear model.

Actions (login required)

View Item