Comparison Forecasting with Double Exponential Smoothing and Artificial Neural Network to Predict the Price of Sugar

Fauziah Nasir Fauziah, FFF (Universitas Nasional & ICT Company, Indonesia)

Forecasting is a method to predict the future using data and the last information as a tool assist planning to be effective and efficient. Research aim to compare forecasting model for double exponential smoothing method and artificial using secondary data price of sugar in weekly by calculating the average price of sugar in seven traditional markets in Depok 2014 to 2016. The program statistics used zaitun time series. The result is using double exponential smoothing Brown Method the value a=0.6 Best model for artificial neural network using 12-8-1. The value of MSE produced by double exponential smoothing method of Brown is 403282 while the Artificial of Neural Network method 15341.2 The value of MAPE using double exponential smoothing method 1.12 while Artificial Neural Network is 0.74. The Conclusion that Artificial Neural Network method is more appropriate the predict forecasting average price of sugar in Depok

Journal: International journal of simulation: systems, science & technology V18

Published: Dec 30, 2017

DOI: 10.5013/IJSSST.a.18.04.13