Volume 17 , Issue 4 , December 2015 , Pages 165-180
Abbas Mohamed Salah 1 ; Tarik Ahmed Rashid 1 ; Shareef Maulod Shareef 1
1 College of Engineering, Salahaddin University
Exchange rate prices play a significant role in economic and financial systems. The
precise forecasting of the exchange rate will aid to predict the situations of other
aspects in the future. This offers useful information for economist traders to fulfill vital
actions to eliminate risks that might lead to financial losses. One vital and common
approach for forecasting is a quantitative technique. The quantitative technique is split
into two approaches; fundamental and technical analysis. Both approaches have one
different, which is the economic factor. The fundamental analysis uses economic
factors while the other not, but both of them have drawbacks. In fundamental
technique, the forecasters cannot understand the economic factors totally, the values
and quantity of the factors unstable, while the technical analysis neglects the factors
totally. This paper suggests a new technique which is appeared to be as a hybridized of
the both techniques for forecasting exchange rate prices to overcome the issues
mentioned above. The proposed technique makes new factors that have real values and
have a real impact on the accurate result. It also derives the inputs and factors in the
prediction models by a feature extraction technique which produces very accurate
results.