Reconstructing Subnational Labor Indicators in Colombia (1993-2025): Datasets

This dataset accompanies the paper "Reconstructing Subnational Labor Indicators in Colombia: An Integrated Machine and Deep Learning Approach". It provides monthly and annual labor indicators for all 33 Colombian departments from 1993 to 2025. The database includes seven key variables: Employment Unemployment Labor force participation (PEA) Inactivity Working-age population (PET) Total population Informality rate The reconstruction framework integrates temporal disaggregation, time-series splicing, statistical and deep learning models, and institutional covariates. Estimates are internally coherent, scaled to demographic projections, and aligned with national benchmarks. Validation against official GEIH departmental aggregates and city-level informality indicators yields in-sample MAPEs below 2.3%, confirming high predictive accuracy. This is the first spatially exhaustive and temporally consistent dataset of monthly labor measures for Colombia, enabling the study of long-term dynamics, regional disparities, and policy design. File format: Excel files (one per entity type and frequency).Coverage: All 33 departments of Colombia, 1993–2025. Citation: Please cite both this dataset and the associated preprint:Vera-Jaramillo, J. (2025). Reconstructing Subnational Labor Indicators in Colombia: An Integrated Machine and Deep Learning Approach. arXiv:2508.12514.