PREDIKSI HARGA SAHAM PERUSAHAAN KELAPA SAWITMENGGUNAKAN PEMODELAN MLP DAN RBF

  • Linda Sari Dewi Sistem Informasi STMIK Nusa Mandiri
Keywords: Multilayer Perceptron, MLP, Prediction, Radial Basis Function, RBF

Abstract

There is great demand estimation method to obtain accurate predictions of the future and use the predictions to the decision making process. This study aims to estimate the daily stock price PT. Astra Agro Lestari Tbk, Indonesia's crude oil company, using one of the promising alternative tool, the Neural Network. Two Neural Network model, namely Multilayer Perceptron (MLP) and Radial Basis Function (RBF). Various combinations of data preprocessing techniques, network topology, functions, and training algorithm is presented in order to obtain the best model. The empirical results show that neural networks Radial Basis Function provides the best forecasting accuracy than Multilayer Perceptron. 

References

G. Zhang, B. E. Patuwo, and M. Y. Hu, "Forecasting with artificial neural networks: the state of the art", International Journal of Forecasting, vol.14, 1998, pp.35-62.

P. H. Sherrod. (2010). DTREG – Predictive modeling software.

Rabunal, J. R., & Dorado, J. (2006). Artificial Neural Network in Real-Life Applications. United States of America: Idea Group Publishing.

Setiawan, W. (2008). Prediksi Harga Saham Menggunakan Jaringan Syaraf Tiruan Multilayer Feedforward Network Dengan Algoritma Backpropagation. Konferensi Nasional Sistem dan Informatika, 108-113.

Sureshkumar. (2012). Performance Analysis of Stock Price Prediction using Artificial Neural Network. Global Journal of Computer Science and Technology, 1-9.

Undang-Undang Republik Indonesia, (1995).

www.finance.yahoo.com
Published
2014-03-15
How to Cite
Dewi, L. (2014). PREDIKSI HARGA SAHAM PERUSAHAAN KELAPA SAWITMENGGUNAKAN PEMODELAN MLP DAN RBF. Jurnal Techno Nusa Mandiri, 11(1), 79-85. Retrieved from http://ejournal.nusamandiri.ac.id/index.php/techno/article/view/173
Article Metrics

Abstract viewed = 6 times
PDF downloaded = 33 times