<![CDATA[One of the main problems faced by policymakers in economic monitoring is the limited availability of predictive tools that can comprehensively and in real time measure economic growth risks, particularly amid financial market volatility and rapid changes in economic indicators. This study aims to nowcast Indonesian economic growth using the Growth at Risk (GaR) approach by applying the Mixed Data Sampling-Quantile Regression (MIDAS-QR) model. This approach predicts economic risks across different quantiles, capturing best- and worst-case scenarios by integrating multi-frequency indicators, namely the Financial Conditions Index (FCI), External Financial Environment Index (EFEI), and Macroeconomic Prosperity Leading Index (MPLI), summarized using Principal Component Analysis (PCA). Prediction accuracy is evaluated using Quantile Mean Absolute Error (QMAE), Quantile Root Mean Squared Error (QRMSE), and Clark-West (CW) test metrics. The analysis utilizes a dataset of Indonesia covering the period from January 2001 to March 2025, combining quarterly GDP growth data as the dependent variable and monthly predictor variables sourced from the Central Statistics Agency (BPS), Bank Indonesia, and the Indonesia Stock Exchange. The findings show that the MIDAS-QR model significantly improves the accuracy of GaR forecasting in Indonesia relative to conventional approaches. It effectively captures risk asymmetries across quantiles, minimizes predictive errors, and facilitates the timely detection of economic downturns, offering valuable insights for early action. This study highlights the strategic role of high-frequency data in enhancing forecast precision and real-time economic risk monitoring in Indonesia. The application of the MIDAS-QR model presents a valuable tool for policymakers in formulating proactive responses to global economic uncertainty and fostering resilient economic growth.]]>
Copyrights © 2026
