On nonlinear dynamic binary time series

Tagore, Vickneswary (2006) On nonlinear dynamic binary time series. Masters thesis, Memorial University of Newfoundland.

[img] [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.

Download (1MB)

Abstract

There are many practical situations where one may encounter binary data over a long period of time. For example, in clinical studies, one may be interested in examining the effects of certain time dependent covariates on the binary asthma status (yes or no) of an individual recorded daily over a few months. The analysis of this type of binary time series data is, however, not adequately addressed in the literature. In the thesis, we review three widely used binary time series models and discuss their advantages and draw-backs mainly with regard to their correlation structures. We then provide inferences for a non-linear conditional dynamic binary model which appears to accommodate correlations with full ranges. -- With regard to the estimation of the regression and a dynamic dependence parameters we use the well-known maximum likelihood (ML) and various versions of the generalized quasilikelihod (GQL) approaches. The relative performances of these approaches are examined through a simulation study. A conditional GQL (CGQL) approach appears to be quite simple and at the same time it produces the same estimates as that of the ML approach. A lag 1 forecasting for a future binary probability is also studied mainly through simulations.

Item Type: Thesis (Masters)
URI: http://research.library.mun.ca/id/eprint/11506
Item ID: 11506
Additional Information: Bibliography: leaves 59-60.
Department(s): Science, Faculty of > Mathematics and Statistics
Date: 2006
Date Type: Submission
Library of Congress Subject Heading: Time-series analysis.

Actions (login required)

View Item View Item

Downloads

Downloads per month over the past year

View more statistics