PREDIKSI HARGA SAHAM MENGGUNAKAN METODE JARINGAN SYARAF TIRUAN BACKPROPAGATION DAN ADAPTIVE NEURO INFERENCE SYSTEM (ANFIS)

Abstract

Determination of the results of the closing price prediction using the backpropagation (ANN) and the Adaptive Neuro Fuzzy Inference System (ANFIS). Backpropagation and ANFIS are methods that are often used in predicting data using historical data on the closure of shares of the Company (Persero) of PT Telekomunikasi Indonesia Tbk (TLKM.JK) recorded based on the number of working days in 1 month 25 days and not including holidays. The data is then processed with Microsoft Excel and processed using Matlab 2013a.
Based on the simulation results of the comparative predictions of Backpropagation and ANFIS. With a MAPE Backpropagation value of 13.44 and ANFIS 9.19. This shows that in this study the ANFIS method has more accurate prediction results.