Kajian Penerapan Artificial Neural Network (ANN) Untuk Memprediksi Harga Saham Mustika Ratu Dengan Metode Support Vector Machine (SVM) dan Multi Layer Perceptron (MLP)


  • Budi Wasito Program Studi Informatika Institut Bisnis dan Informatika Kwik Kian Gie


Every investor who transact in the capital markets hope benefits. But the stock has characteristics of
high risk-high return, it means that stocks allows investors to make a profit (capital gain) in large
quantities in a short time, but it can also make stock investors suffered heavy losses in a short time.
Investors require a number of methods in an effort to assist the purchase of shares of investment
Data of Stock price is a time series of data in a given period has a unique pattern. Then using machine
learning methods, this research is reviewing the use of SVM and MLP-related objects in PT Mustika
Ratu. Input variable is in the form of historical stock prices from 2007 to 2013. This study tried to
reveal the level of RMSE (Root Mean Square Error) between SVM and MLP. Concluded that the
method of learning by using the MLP has a lower RMSE than using SVM.





Vol 2, No 1 (2013)