Predicting Indonesian Inflation Rate Using Long Short-Term Memory (LSTM)

Wijaya, Muhammad Krisna
Nastiti, Faulinda Ely
Farida, Anisatul

Journal Of Artificial Intelligence And Software Engineering
Vol 5, No 3 (2025): September

Inflation is a crucial economic indicator that requires an accurate prediction model. This research aims to develop a prediction system for the monthly inflation rate in Indonesia using the Long Short-Term Memory (LSTM) architecture. The method includes historical data acquisition from Bank Indonesia, preprocessing with Min-Max Scaler normalization, and training a univariate LSTM model. Evaluation results show excellent performance with an MAE of 0.2999, an RMSE of 0.3903, and an R² of 0.8796, indicating the model explains 88% of the data's variability. It is concluded that LSTM is effective for inflation forecasting in Indonesia and serves as a solid baseline for future research.