PREDIKSI INFLASI INDONESIA DENGAN MODEL ARTIFICIAL NEURAL NETWORK
This research examines and analyzes the use of Artificial Neural Networks (ANN) asa forecasting tool. Specifically a neural networkâ€™s ability to predict future trends ofinflation is tested. Accuracy is compared against a traditional forecasting method,multiple linear regression analysis. Finally, the probability of the modelâ€™s forecastbeing Â correct Â is Â calculated Â using Â conditional Â probabilities. While Â only Â brieflydiscussing Â neural Â network Â theory, Â this Â research Â determines Â the Â feasibility Â andpracticality of using neural networks as a forecasting tool for inflation in Indonesia.This study builds upon the work done by Edward Gately in his book Neural Networksfor Financial Forecasting. This research validates the work of Gately and describesthe Â development of Â a Â neural network Â that Â achieved an Â 86 Â percent probability Â ofpredicting an Â inflation Â rise, while multiple Â regression Â analysis Â is only Â to predictinflation that achieved a 16%. Â It was concluded that neural networks do have Â thecapability to forecast inflation and, if properly trained, we could benefit from the useof this forecasting tool.
Keywords: neural networks, Â inflation, Â time Â series analysis, Â forecasting, artificialintelligence