APPLICATION OF ARTIFICIAL NEURAL NETWORK ON FORECASTING EXCHANGE RATE OF RUPIAH TO US DOLLAR

Yogo Dwi Prasetyo

Abstract


Abstract: Neural network is one of the technologies in the field of Artificial Intellegence in which data pattern identification of the Rupiah forecasting system to US dollar can be done by the method of learning approach. Based on its learning ability, the neural network can be trained to study and analyze the past data patterns then find a formula or function that will correlate the past data pattern with desired output. This study aims to explain the mechanism of forecasting using artificial neural networks and to determine the neural network model in forecasting Rupiah against US dollar. The exerted software to create the exchange rate forecasting program is Matlab 7.0. The best model of simulation results shown in architecture with 10 input layers, 1 hidden layer with 2 hidden neurons, 1 output layer, backpropagation training algorithm, and bipolar sigmoid activation function. The results of calculations using artificial neural networks obtained MAE values Rp 45.74 in the data testing.

 

Keywords:Forecast, Exchange Rate, Artificial Neural Networks, Backpropagation Algorithm.


Full Text:

PDF

References


Bishop, C.M. 1995. Neural Networks for Pattern Recognition. Oxford University Press.

Kecman, Vojislav. 2001. Learning anf Softcomputing: Support Vector Machines. Neural Networks and Fuzzy Logic Models. USA. Massachusetts Institute of Technology.

Kusumadewi, Sri. 2004. Membangun Jaringan Syaraf Tiruan Menggunakan MATLAB & EXCEL LINK. Yogyakarta: Graha Ilmu.

Nuraeni, Ai. 2009. Feed Forward Neural Network untuk Small Area Estimation pada Kasus Kemiskinan. [Tesis]. Surabaya : ITS.

Siang, Jong Jek. 2005. Jaringan Syaraf Tiruan dan Pemrogramannya Menggunakan Matlab. Yogyakarta: Penerbit Andi

Suhartono. 2007. Feed Forward Neural Networks untuk Pemodelan Runtun Waktu [Disertasi]. Yogyakarta :UGM.

Suryono, Hady. 2009. Pemodelan Auto Regressive Integrated Moving Average With Exogens Factor-Neural Network (ARIMAX-NN) pada Data Inflasi Indonesia. [Tesis]. Surabaya : ITS


Article Metrics

Abstract view : 293 times
PDF - 129 times

Refbacks

  • There are currently no refbacks.