Review Articles

Model-based small area estimation with no samples within the areas, by benchmarking to marginal cross-sectional and time-series estimates

Danny Pfeffermann ,

a National Statistician and CBS Director, Jerusalem, Israel;b Department of Statistics, Hebrew University of Jerusalem, Jerusalem, Israel;c Southampton Statistical Sciences Research Institute, University of Southampton, Southampton, UK

Michael Sverchkov ,

d Bureau of Labor Statistics, Washington, DC, USA

Richard Tiller ,

d Bureau of Labor Statistics, Washington, DC, USA

Lizhi Liu

d Bureau of Labor Statistics, Washington, DC, USA

Pages 28-42 | Received 25 Jan. 2019, Accepted 19 Jan. 2020, Published online: 31 Jan. 2020,
  • Abstract
  • Full Article
  • References
  • Citations


Official monthly U.S. labour force estimation at the sub-State level (mostly counties) is based on what is known as the ‘Handbook’ (HB) method, one of the earliest uses of administrative data for small area estimation. The administrative data, however, are poor in coverage and have conceptual deficiencies. Past attempts to correct for the resulting bias of the HB estimates by informal (implicit) modelling have not been successful, due to the absence of regular direct monthly survey estimates at the sub-State level. Benchmarking the sub-State HB estimates each month to the State model dependent estimates helps to correct for an overall bias, but not in individual areas. In this article we propose benchmarking additionally to the annual model-dependent area estimates. The annual models include known administrative data as covariates, and are used to define corresponding monthly sub-State models, which in turn enable producing monthly synthetic estimates as possible substitutes for the HB estimates in real time production. Variance estimates, which account for sampling errors and the errors of the model dependent estimators are developed. Data for sub-State areas in the State of Arizona are used for illustration. Although the methodology developed in this article stems from a particular (but very important) application, it is general and applicable to other similar problems.


  1. Bare, S. (2012, August 15). Yuma’s paradox: Despite rising unemployment, population still grows. Cronkite News. [Google Scholar]
  2. Bikker, R., Daalmans, J., & Mushkudiani, N. (2013). Benchmarking large accounting frameworks: A generalized multivariate model. Economic Systems Research25, 390–408. doi: 10.1080/09535314.2013.801010 [Taylor & Francis Online][Web of Science ®], [Google Scholar]
  3. Deming, W. E., & Stephan, F. F. (1940). On a least squares adjustment of a sample frequency table when the expected marginal totals are known. Annals of Mathematical Statistics11, 427–444. doi: 10.1214/aoms/1177731829 [Crossref], [Google Scholar]
  4. Denton, F. T. (1971). Adjustment on monthly or quarterly series to annual totals: An approach based on quadratic minimization. Journal of the American Statistical Association66, 99–102. doi: 10.1080/01621459.1971.10482227 [Taylor & Francis Online][Web of Science ®], [Google Scholar]
  5. Deville, J. C., & Särndal, C.-E. (1992). Calibration estimators in survey sampling. Journal of the American Statistical Association87, 376–382. doi: 10.1080/01621459.1992.10475217 [Taylor & Francis Online][Web of Science ®], [Google Scholar]
  6. Dostal, L., Gabler, S., Ganninger, M., & Münnich, R. (2016). Frame correction modeling with applications to the German register-assisted census 2011. Scandinavian Journal of Statistics43, 904–920. doi: 10.1111/sjos.12220 [Crossref][Web of Science ®], [Google Scholar]
  7. Harvey, A. C. (1989). Forecasting structural time series with the Kalman filter. Cambridge, UK: Cambridge University Press. [Google Scholar]
  8. Isaki, C. T., Tsay, J. H., & Fuller, W. A. (2000). Weighting sample data subject to independent controls. Survey Methodology30, 35–44. [Google Scholar]
  9. Kromer, B., & Howard, D. (2011). Comparison of ACS and CPS data on employment status (Working Paper Number SEHSD-WP2011-31). Social, Economic and Housing Statistics Division, U.S. Census Bureau. [Google Scholar]
  10. Nagaraja, C., & McElroy, T. (2015). On the interpretation of multi-year estimates of the American community survey as period estimates. Journal of the International Association of Official Statistics31, 661–676. [Google Scholar]
  11. Pfeffermann, D., & Barnard, C. H. (1991). Some new estimators for small area means with application to the assessment of farmland values. Journal of Business & Economic Statistics9, 73–83. [Taylor & Francis Online][Web of Science ®], [Google Scholar]
  12. Pfeffermann, D., & Tiller, R. (2006). State-space modelling with correlated measurement errors with application to small area estimation under benchmark constraints. Journal of the American Statistical Association101, 1387–1397. doi: 10.1198/016214506000000591 [Taylor & Francis Online][Web of Science ®], [Google Scholar]
  13. Purcell, N. J., & Kish, L. (1980). Postcensal estimates for local areas (or domains). International Statistical Review48, 3–18. doi: 10.2307/1402400 [Crossref][Web of Science ®], [Google Scholar]
  14. Rao, J. N. K., & Molina, I. (2015). Small area estimation (2nd ed.). Hoboken, NJ: Wiley. [Crossref], [Google Scholar]
  15. Sverchkov, M., & Tiller, R. (2016). Calibration on partly known counts in frequency tables with application to real data. 2016 JSM meetings, proceedings of the section on survey methods research, Chicago (pp. 711–713). [Crossref], [Google Scholar]
  16. Tiller, R. B. (1992). Time series modeling of sample survey data from the U.S. Current population survey. Journal of Official Statistics8, 149–166. [Google Scholar]
  17. U.S. Bureau of Labor Statistics. (1960). Handbook on estimating unemployment (BES No. R-185, U.S.) Department of Labor, Bureau of Employment Security. [Google Scholar]
  18. U.S. Bureau of Labor Statistics, Handbook of Methods. (2019). Local area unemployment statistics. Retrieved from [Google Scholar]
  19. Zhang, L.-C., & Chambers, R. L. (2004). Small area estimates for cross-classifications. Journal of the Royal Statistical Society, Series B66, 479–496. doi: 10.1111/j.1369-7412.2004.05266.x [Crossref], [Google Scholar]