Comparison of neural network and hybrid genetic algorithm-neural network in forecasting of Philippine peso-US dollar exchange rate

• Main Authors: Torregoza, Mark Lorenze R., Dadios, Elmer P.
• Format: text
• Published: Animo Repository 2014
• Subjects: Foreign exchange rates—Forecasting; Neural networks (Computer science); Evolutionary computation; Finance and Financial Management; Manufacturing
• Online Access: https://animorepository.dlsu.edu.ph/faculty_research/1913
• Institution: De La Salle University

This paper presents a new method in forecasting Philippine Peso to US Dollar exchange rate. Compared to the conventional way, in which the Philippine Dealing System (PDS), as monitored by the Central Bank, determines the rate by analysing demand and supply, the use of artificial neural network, having consumer price index, inflation rate, lending interest rate and purchasing power of the peso as the inputs is presented in this paper. Though foreign exchange rates vary on a daily basis, the output of this paper is prediction of the average foreign exchange rate every month. Artificial Neural Network serves as a powerful tool in forecasting Philippine Peso to US Dollar exchange rate not requiring expert knowledge in banking and finance thus letting the public gain access to a helpful beacon which is the foreign exchange rate. However, the accuracy of the forecast using artificial neural network is highly dependent on the volume of the training data, in this paper, an alternative algorithm that will increase the accuracy of the conventional artificial neural network with limited volume of training data is presented and analyze. © 2014 IEEE.