Using Small Area Estimation to Produce Official Statistics

The USDA National Agricultural Statistics Service (NASS) and other federal statistical agencies have used probability-based surveys as the foundation for official statistics over half a century. Non-survey data that can be to improve accuracy precision of estimates such administrative, remotely sensed, retail become increasingly available. Both frequentist Bayesian models are combine survey non-survey in principled manner. NASS has recently adopted subarea three its national programs: farm labor, crop county estimates, cash rent estimates. Each program provides valuable at multiple scales geography. For each program, technical challenges had met strenuous review completed before could statistics. Moving out research phase into production required major changes process cultural shift. With implemented models, now measures uncertainty, transparency, reproducibility