Empirical modelling of high-frequency foreign exchange rates

There is a wealth of information available on modelling foreign exchange time series data, however, research studies on modelling and predicting high frequency foreign exchange data is less prominent. Furthermore, there does not appear to be much evidence supporting work on the modelling and prediction of high frequency South African Rand/United States Dollar (ZAR/USD) exchange rates. A fair amount of noise is embedded in high frequency time series data, especially the ZAR/USD exchange rates, and the modelling of these time series requires the use of specialized models. In addition, lengthy high frequency foreign exchange data is largely unavailable for the South African market. This dissertation undertakes empirical explorations to model high frequency foreign exchange time series (primarily the ZAR/USD time series), through the use of multi-agent neural networks, linear Kalman filters and fuzzy Markov chain theory.

Grantor dc:publisher.institution

Department of Mathematics and Applied Mathematics

Year dc:date.issued

2004

Author dc:creator

• Packirisamy, Someshini

Advisor dc:contributor.advisor

• Guo, Renkuan

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/11427/5963

OAI identifier oai:identifier

oai:open.uct.ac.za:11427/5963