PERAMALAN HARGA MINYAK MENTAH DUNIA MENGGUNAKAN METODE RADIAL BASIS FUNCTION NEURAL NETWORK

Abstract

Oil is the most important commodity in everyday life, because oil is one of the main source of energy that is needed for the people. Changes in crude oil prices greatly affect the economic conditions of a country. To forecast crude oil prices, the past data of the crude oil that is the time series data will be studied so that will produce crude oil price forecast in the future. Model of Radial Basis Function Neural Network is suitable for large-scale data processing, because this model does not require the use of all data input and has a total processing time of rapid system. This model has a network architecture in the form of input layer, hidden layer and output layer. Analysis conducted on the data as much as 1286 taken as 100 the data thus obtained value of 0.9145 MSE training and training MAPE value of 0.74%, while for the testing of 4.2739 MSE and MAPE testing value is 1.63%. Based on the results of forecasting, crude oil prices on July 29, 2015 until August 2, 2015 at USD $ 55.91 per barrel. Keywords: Radial Basis Function Neural Network (RBFNN), Time Series, Crude Oil, MSE, MAPE, Forcasting