Forecasting Indonesia’s Inflation-at-Risk Using Vector Autoregression with Exogenous Variables Model

Authors

  • Ni Kadek Juliarini Department of Statistics, Institut Teknologi Sepuluh Nopember, Surabaya 60111, INDONESIA
  • Dedy Dwi Prasty Department of Statistics, Institut Teknologi Sepuluh Nopember, Surabaya 60111, INDONESIA https://orcid.org/0000-0003-1194-769X
  • Muhammad Sjahid Akbar Department of Statistics, Institut Teknologi Sepuluh Nopember, Surabaya 60111, INDONESIA

DOI:

https://doi.org/10.22452/mjs.vol45sp2.10

Keywords:

IaR, FCI, FSI, VARX, Forecasting

Abstract

Inflation is a critical economic indicator influencing stability and monetary policy, making accurate forecasting essential. This study employs the Vector Autoregression with Exogenous Variables (VARX) model to simultaneously forecast Indonesia's inflation across various sectors, including food, health, housing, clothing, and education, incorporating the Financial Condition Index (FCI) and Financial Stress Index (FSI). The FSI evaluates financial stress using data from banking, foreign exchange, stock markets, and government securities, while the FCI assesses economic conditions through Principal Component Analysis (PCA). VARX is rarely used in this context and differs from previous studies in that it focuses on the risk of extreme inflation. In addition, this study introduces the use of the Correlation Information Criterion (CIC) as an evaluation tool, which has not been widely applied in similar research. The fluctuation patterns of each sectoral inflation generally display distinct characteristics, enabling a comprehensive evaluation of the VARX model’s ability to forecast data with varying fluctuation patterns. This approach differs from previous studies, which typically utilize data with uniform fluctuation patterns. Furthermore, the Seasonal Trend Decomposition Based on Loess (STL) method is employed to mitigate seasonal effects in the data and improve forecasting accuracy. This technique is rarely applied in this context. Analyzing monthly data from February 2010 to December 2024, the VARX(1) model effectively forecasts Indonesia's Inflation-at-Risk (IaR) with high accuracy, as measured using the Symmetric Mean Absolute Percentage Error (SMAPE) criterion. Seasonal inflation in the health sector is addressed using the Seasonal Trend Decomposition Based on Loess (STL) method, which further improves forecasting precision. However, the model struggles with extreme fluctuations, highlighting the need for future research into more advanced predictive models for volatile economic conditions.

References

Akbar, M. S., Setiawan, S., Suhartono, S., Ruchjana, B. N., Prastyo, D. D., Muhaimin, A., & Setyowati, E. (2020). A Generalized Space-Time Autoregressive Moving Average (GSTARMA) Model for Forecasting Air Pollutant in Surabaya. J. Phys.: Conf. Ser. 1490 (012022).

Chicco, D., Warrens, M. J., & Jurman, G. (2021). The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation. PeerJ Computer Science, 18.

Cleveland, R. B., Cleveland, W. S., McRae, J. E., & Terpenning, I. (1990). STL: A Seasonal-Trend Decomposition Procedure Based on Loess. Journal of Official Statistics, 6(1), 3-73.

Farih, I., & Prastyo, D. D. (2022). Forecasting Electricity Consumption Based on Economics and Social Indicators Using VAR Model with Exogenous Variable: Evidence From East Java Province. in 022 6th International Conference on Information Technology, Information Systems and Electrical Engineering (ICITISEE). Yogyakarta.

Ferrara, L., Mogliani, M., & Sahuc, J.-G. (2022). High-frequency monitoring of growth at risk. International Journal of Forecasting, 38, 582-595.

Gujarati, D. N. (2003). Basic Econometrics. New York: Gary Burke.

Hikmawati, F., Suhartono, S., & Prastyo, D. D. (2021). A Hybrid GSTARX-Jordan RNN Model for Forecasting Space-Time Data with Calendar Variation Effect. J. Phys.: Conf. Ser. 1752 (012013).

Juhro, S. M., & Iyke, B. N. (2019). Monetary Policy and Financial Condition in Indonesia. Bulletin of Monetary Economics and Banking, 21, 283-302.

Novianto, M. A., Suhartono, S., Prastyo, D. D., Suharsono, A., & Setiawan, S. (2018). GSTARIX Model for Forecasting Spatio-Temporal Data with Trend, Seasonal and Intervention. J. Phys.: Conf. Ser. 1097 (012076).

Ocampo, S., & Rodríguez, N. (2012). An Introductory Review of a Structural VAR-X Estimation and Applications. Revista Colombiana de Estadística, 35, 479-508.

Park, C.-Y., & Mercado, R. V. (2014). Determinants of financial stress in emerging market economies. Journal of Banking & Finance, 45, 199-224.

Prastyo, D. D., Nabila, F. S., Suhartono, S., Lee, M. H., & Suhermi, N. F.-F. (2018). VAR and GSTAR-Based Feature Selection in Support Vector Regression for Multivariate Spatio-Temporal Forecasting. In Soft Computing in Data Science (SCDS), Communications in Computer and Information Science, vol 937. Springer.

Safuan, S., Sugandi, E. A., Aziz, O. Q., & Triandhari, R. (2022). Indonesia's Financial Stress Events and Macroeconomic Dynamics. Bulletin of Monetary Economics and Banking, 25, 323-370.

Sesay, A., Suhartono, S., & Prastyo, D. D. (2020). Forecasting Exchange Rate Across Countries with Gold Price as Exogenous Variable Using Transfer Function and VARI-X Model. MATEMATIKA, 36(3), 181-196.

Suhartono, S., Gazali, M. M., & Prastyo, D. D. (2018). VARX and GSTARX Models for Forecasting Currency Inflow and Outflow with Multiple Calendar Variations Effect. MATEMATIKA, 34(3), 57–72.

Suhartono, S., Prastyo, D. D., Kuswanto, H., & Lee, M. H. (2018). Comparison between VAR, GSTAR, FFNN-VAR and FFNN-GSTAR Models for Forecasting Oil Production. MATEMATIKA, 34(1), 103–111.

Tsay, R. S. (2002). Analysis of Financial Time Series. Canada: John Wiley & Sons, Inc.

Zivot, E., & Wang, J. (2003). Modeling Financial Time Series with S-Plus. New York: Springer Science+Bu siness Media.

graphical abstract

Downloads

Published

15-11-2026

Most read articles by the same author(s)