• Login
    View Item 
    •   USU-IR Home
    • Faculty of Computer Science and Information Technology
    • Department of Information Technology
    • Master Theses
    • View Item
    •   USU-IR Home
    • Faculty of Computer Science and Information Technology
    • Department of Information Technology
    • Master Theses
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Analisis Accelarated Learning pada Backpropagation dalam Peramalan Indeks Harga Saham Gabungan

    View/Open
    Fulltext (1.946Mb)
    Date
    2013
    Author
    Barus, Armansyah
    Advisor(s)
    Zarlis, Muhammad
    Budhiarti, Erna
    Metadata
    Show full item record
    Abstract
    Forecasting stock is much needed equity investor in deciding when to sell and buy a stock index. Metematis many theories that have been used to get the right results but forecasting system that is often used is static. For cases whose value is dynamic, it is difficult in the development of mathematical models. In accordance with the development of computer technology, the application of Artificial Neural Network method becomes easier to model the dynamic system. Resilient Backpropagation is one model of Artificial Neural Network (ANN), which has been implemented for forecasting stock index. Risilient Backpropagation learning has the ability to output data and predict the future based on the learning outcomes that have been done. In this study, ANN method will be applied to forecast price of the Open, High, Low and Close in the Indonesia Stock Exchange stock index. With the data in the form of daily stock prices, artificial neural networks are designed to give the weights that are used to predict the stock price in the next day. In this study, we can conclude that the network uses the input variable opening price, high, low and close and output variables open, high, low and close using the two (2) hidden layer with the number of nodes and the number 100 hidden layer1 layer2 150 hidden nodes produce the best forecasting accuracy. This research is still in its early stages, where there are many factors that can be developed or researched further.
     
    Peramalan saham merupakan hal yang sangat dibutuhkan investor saham dalam menentukan kapan harus menjual dan membeli suatu indeks saham. Banyak teori metematis yang telah digunakan untuk mendapatkan hasil peramalan yang tepat tetapi system peramalan yang sering digunakan masih statis. Untuk kasus yang nilainya dinamis, sangatlah sulit dalam pengembangan model matematisnya. Sesuai dengan perkembangan teknologi komputer, penerapan metode Artificial Neural Network menjadi lebih mudah dalam memodelkan system dinamis. Resilient Backpropagation adalah salah satu model Artificial Neural Network (ANN) yang telah diimplementasikan untuk peramalan indeks saham. Risilient Backpropagation mempunyai kemampuan untuk melakukan pembelajaran dan meramalkan data keluaran pada waktu mendatang berdasarkan hasil pembelajaran yang telah dilakukan. Pada penelitian ini, metode ANN akan diterapkan untuk meramalkan harga Open, High, Low dan Close dalam indeks saham Bursa Efek Indonesia. Dengan data berupa harga saham harian, jaringan syaraf tiruan yang dirancang akan menghasilkan bobot-bobot yang digunakan untuk meramal harga saham di hari berikutnya. Dalam penelitian ini, dapat disimpulkan jaringan yang menggunakan variabel input harga opening, high, low dan close dan variabel output open, high, low dan close dengan menggunakan dua (2) hidden layer dengan jumlah node hidden layer1 100 dan jumlah node hidden layer2 150 menghasilkan keakuratan peramalan yang paling baik. Penelitian ini masih dalam tahap awal, dimana masih banyak faktor yang dapat dikembangkan atau diteliti lebih lanjut.

    URI
    http://repositori.usu.ac.id/handle/123456789/40579
    Collections
    • Master Theses [621]

    Repositori Institusi Universitas Sumatera Utara (RI-USU)
    Universitas Sumatera Utara | Perpustakaan | Resource Guide | Katalog Perpustakaan
    DSpace software copyright © 2002-2016  DuraSpace
    Contact Us | Send Feedback
    Theme by 
    Atmire NV
     

     

    Browse

    All of USU-IRCommunities & CollectionsBy Issue DateTitlesAuthorsAdvisorsKeywordsTypesBy Submit DateThis CollectionBy Issue DateTitlesAuthorsAdvisorsKeywordsTypesBy Submit Date

    My Account

    LoginRegister

    Repositori Institusi Universitas Sumatera Utara (RI-USU)
    Universitas Sumatera Utara | Perpustakaan | Resource Guide | Katalog Perpustakaan
    DSpace software copyright © 2002-2016  DuraSpace
    Contact Us | Send Feedback
    Theme by 
    Atmire NV