Possibility of predicting inflation: Using machine learning model
Tegshjargal Sodnomdavaa1
, Tsolmon Sodnomdavaa2* , Namuun Amgalanbat3
*Corresponding Author: Tsolmon Sodnomdavaa
1School of Management, Mongolian University of Science and Technology, Mongolia, tegshjargal@gmail.com
2*School of Engineering and Economics, Mandakh University, Mongolia, tsolmon@mandakh.edu.mn
3Business Development Department, Max Group, Mongolia, amgalanbatbadmaa@gmail.com
Digital Object Identifier: https://doi.org/10.53468/mifyr.2025.05.02.49
Abstract: In recent years, global inflationary pressures have intensified due to multiple risk factors such as the COVID-19 pandemic, disruptions in global supply chains, geopolitical conflicts, and trade wars. These macroeconomic shocks have increased the need for accurate and timely inflation forecasting, particularly in emerging economies. This study explores the application of machine learning techniques for forecasting inflation in Mongolia. Using monthly data from January 2004 to January 2025-including 6,325 observations covering the inflation rate and 25 macroeconomic indicators, the study compares the predictive performance of five models: XGBoost, Random Forest, Quantile Regression, SARIMA, and GARCH. The dataset was divided into training (70%) and testing (30%) sets, and models were evaluated using R², RMSE, and MAE. Among the models, XGBoost achieved the highest accuracy (R² = 0.88, RMSE = 0.12, MAE = 0.11), demonstrating its effectiveness in capturing complex economic dynamics. In contrast, traditional models like SARIMA and GARCH were less suitable for long-term forecasting. The findings underscore the potential of machine learning methods, particularly XGBoost, to support data-driven economic policy and inflation targeting in Mongolia.
Keywords–Inflation forecasting; Machine learning; XGBoost; Economic shocks; Time series
Article History: Received 20 May 2025, Received in revised form 22 May 2025, Accepted 9 June 2025
Download file: httpsdoi.org10.53468mifyr.2025.05.02.49