VIIRS nighttime lights in the estimation of cross-sectional and time-series GDP

Abstract

This study extends previous applications of DMSP OLS nighttime lights data to examine the usefulness of newer VIIRS lights in the estimation of economic activity. Focusing on both US states and metropolitan statistical areas (MSAs), we found that the VIIRS lights are more useful in predicting cross-sectional GDP than predicting time-series GDP data. This result is similar to previous findings for DMSP OLS nighttime lights. Additionally, the present analysis shows that high-resolution VIIRS lights provide a better prediction for MSA GDP than for state GDP, which suggests that lights may be more closely related to urban sectors than rural sectors. The results also indicate the importance of considering biases that may arise from different aggregations (the modifiable areal unit problems, MAUP) in applications of nighttime lights in understanding socioeconomic phenomenon.