Digital twinUS & West
Evaluating workforce training programs with tax records and experimental benchmarks
Using anonymized income tax records for all US adults from 1999 to 2025, the study estimates earnings effects of workforce training programs, predicting counterfactuals with machine learning from work histories, demographics and location. Estimated bias is benchmarked against 37 randomized controlled trials.
Why it matters
It illustrates how large administrative data and learned models can scale program-effect estimation while being calibrated against experimental benchmarks.
Relevant to DT products
- Original headline
- Experimentally Validated Digital Twins: Evaluating Workforce Training Programs in the Age of Big Data
Translation and summary are machine-generated from the source. The full article is not republished; its rights belong to the source.