YILDIZ, ÖMÜR
Master Thesis – 2006
Business Administration
Advisor: Birol YILDIZ, Assistant Professor
The purpose of this study was to perform an application of
Artificial Neural
Networks (ANN) which are one of the promising Artificial Intelligence techniques
nowadays, on economical and financial sectors.
Data were collected from the official web site of Central Bank of the Republic
of Turkey. 41 economical
parameters and variables which affect the fluctuations of
the USD were used in this study. 1867 days of data between 4 January 1999 and 28
February 2006 used in this study.
As a result of this study we determined how affects these parameters such as
the structure of ANN, number of layers in the ANN and artificial neurons in the
layers, the training methods to reach the best solution. We have decided that ANNs
can be used as a supplementary tool to predict future value and trend of the USD.