BANKING PRICE FORECASTING APPLICATION USING NEURAL NETWORK TIME SERIES METHOD

DIANA NUR AROFAH, FARDANTO SETYATAMA, WIWIET HERULAMBANG

Abstract


In the capital market is a meeting place for investors to make an offer with demand for securities as a means of business funding or as a means for companies to get funds. One of the assets to invest in the capital market is stocks. In terms of business aspects, stock investment has good growth but this does not apply to all stock sectors. Because in fact the development of capital markets in Indonesia turned out to be ups and downs. It can cause changes in demand and supply that will affect investor psychology in predicting stock prices. This stock price forecasting system will be created using the Neural Network Time Series method. Using historical data as a reference in the neural network training process can be used as a basis for predicting bank stock prices the next day. In the tests that have been carried out using the application forecasting stock prices of state banks using the neural network time series method with the backpropagation algorithm, the average accuracy rate of the State Savings Bank (BTN) is 97.32%, Bank Negara Indonesia (BNI) 98.25%, Bank Mandiri 97.68% %, and at Bank Rakyat Indonesia (BRI) 98.59%.


Full Text:

PDF

References


Render and Hizer, Prinsip-Prinsip Manajemen Operasi. Jakarta: Salemba Empat, 2004.

S. Kusumadewi, Artificial Intellegence (Teknik dan Aplikasinya). Jakarta: Graha Ilmu, 2003.

M. Kurniawan, Supervised Neural Networks dan Aplikasinya. Yogyakarta: Graha Ilmu, 2006.

S. Makridakis and S. . Wheelwright, Metode dan Aplikasi Peramalan. Terjemahan Hari Suminto. Jakarta: Binarupa Aksara, 1999.


Refbacks

  • There are currently no refbacks.