Abstract

In emerging economies, where timely quarterly indicators of industrial activity are often limited, costly to compile, and subject to reporting lags, satellite-based measures offer a promising complementary source of information for higher-frequency economic monitoring. Satellite observations provide consistent spatial coverage, are available at daily frequency, and are not constrained by survey response burdens or administrative delays, making them particularly valuable in data-constrained settings. This study explores the use of nighttime satellite radiance as a high-frequency proxy for Indonesia’s manufacturing sector performance, drawing on data from NASA’s VIIRS VNP46A2 sensor and focusing on the ten largest industrial complexes. We construct a composite radiance index from daily observations over the period 2013–2025 to approximate industrial activity.
Employing regression-based analysis, we examine the relationship between the radiance index and quarterly manufacturing GDP growth rates sourced from official statistics. Results reveal a moderate correlation, with the regression model yielding a statistically significant positive coefficient, indicating that variations in nighttime radiance explain a substantial portion of manufacturing sector dynamics.
This approach offers a cost-effective, near-real-time complementary statistic to conventional survey-based indicators, enhancing economic forecasting and policy analysis in data-constrained environments. Future extensions could incorporate alternative model with machine learning for refined spatial disaggregation