PERAMALAN HARGA SAHAM MENGGUNAKAN RECURRENT NEURAL NETWORK DENGAN ALGORITMA BACKPROPAGATION THROUGH TIME (BPTT)

作者: Setiawardhana Setiawardhana , Linda Aqnes Desi Susanti , Arna Fariza

DOI:

关键词: Artificial neural networkStock price forecastingSet (abstract data type)Backpropagation through timeRecurrent neural networkStock exchangeMachine learningNetwork architectureGraph (abstract data type)Computer scienceArtificial intelligence

摘要: In this final project will be made an application to stock price forecasting using RNN - BPTT. modeling the input data is Close Price of shares on Indonesia Stock Exchange (BEI). Then these time series with neural network algorithms recurrent that BPTT algorithm where architecture used Jordan's RNN. Backpropagation Through Time (BPTT) a fairly popular training for network. there are several feedback loops in connection graph. The main concept spread by putting same copy and manage connections back get between next set. To produce accurate forecasting, parameters tested, such as learning rate, number neurons data. We make expected help investors predict fluctuations so they able determine investment policy future good results. Keywords: Forecasting Series, Price, Recurrent Neural Network,

参考文章(8)
A.G. Parlos, O.T. Rais, A.F. Atiya, Multi-step-ahead prediction using dynamic recurrent neural networks Neural Networks. ,vol. 13, pp. 765- 786 ,(2000) , 10.1016/S0893-6080(00)00048-4
Takaki Makino, Proto-predictive representation of states with simple recurrent temporal-difference networks Proceedings of the 26th Annual International Conference on Machine Learning - ICML '09. pp. 697- 704 ,(2009) , 10.1145/1553374.1553464
P.J. Werbos, Backpropagation through time: what it does and how to do it Proceedings of the IEEE. ,vol. 78, pp. 1550- 1560 ,(1990) , 10.1109/5.58337
Barak A. Pearlmutter, Dynamic recurrent neural networks Carnegie Mellon University. ,(1990)
Inés M. Galván, Pedro Isasi, Multi-step Learning Rule for Recurrent Neural Models: An Application to Time Series Forecasting Neural Processing Letters. ,vol. 13, pp. 115- 133 ,(2001) , 10.1023/A:1011324221407
Paolo Tenti, Forecasting Foreign Exchange Rates Using Recurrent Neural Networks Applied Artificial Intelligence. ,vol. 10, pp. 567- 582 ,(1996) , 10.1080/088395196118434