Synthetic Data: A Look Back and A Look Forward

When initially proposed, synthetic data for disclosure control was generally dismissed as unlikely to be implemented in practice. Thirty years later, synthetic data are becoming a staple of the disclosure limitation toolkit. We now see synthetic public use files for several major data products with more on the way. In this article, I review the progression of synthetic data, describe some unresolved challenges, and speculate on its future.

Duke Scholars

Author Jerome P. Reiter Statistical Science