Prediksi Laju Inflasi di Sumatera Utara dengan Metode High Order Intuitionistic Fuzzy Time Series

Abstract

Inflation is one of the key indicators used to assess the economic stability of a region. Uncontrolled inflation rates can negatively impact investment, business activities, and public purchasing power. Therefore, accurate inflation forecasting is essential to support the formulation of effective economic policies. This study aims to forecast the inflation rate in North Sumatra using the High Order Intuitionistic Fuzzy Time Series (HOIFTS) method. This method integrates the concepts of intuitionistic fuzzy sets and high-order time series modeling, enabling it to capture more complex temporal relationships and accommodate uncertainty in historical data. Based on the evaluation of model training data from the 2021–2024 period, the HOIFTS method produced a Mean Absolute Percentage Error (MAPE) of 3.55%, which falls into the category of very high accuracy. Furthermore, the prediction for January 2025 shows that the HOIFTS method was able to generate a forecast value that closely matches the actual data, with a Percentage Error (PE) of 1.69%. These findings indicate that the HOIFTS method not only provides high accuracy during the model training phase but also demonstrates reliable predictive performance, making it a promising approach for supporting future economic decision-making.