A single-layer GRU model using NASA Black Marble nightlights nowcasts Italian municipal taxable income with 1.07 million euro median error (4% of median), statistically outperforming persistence, fixed effects, ARDL, and spatial econometric benchmarks out-of-sample on 2020-2021.
PLoS ONE , volume =
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
econ.EM 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Nowcasting Italian Municipal Income with Nightlights: A Deep Learning Approach
A single-layer GRU model using NASA Black Marble nightlights nowcasts Italian municipal taxable income with 1.07 million euro median error (4% of median), statistically outperforming persistence, fixed effects, ARDL, and spatial econometric benchmarks out-of-sample on 2020-2021.