ABSTRACT

In this article, a new unit level model based on a pairwise penalised regression approach is proposed for problems in small area estimation (SAE). Instead of assuming common regression coefficients for all small domains in the traditional model, the new estimator is based on a subgroup regression model which allows different regression coefficients in different groups. The alternating direction method of multipliers (ADMM) algorithm is used to find subgroups with different regression coefficients. We also consider pairwise spatial weights for spatial areal data. In the simulation study, we compare the performances of the new estimator with the traditional small area estimator. We also apply the new estimator to urban area estimation using data from the National Resources Inventory survey in Iowa.

References

1. Battese, G. E., Harter, R. M., & Fuller, W. A. (1988). An error-components model for prediction of county crop areas using survey and satellite data. Journal of the American Statistical Association, 83(401), 28–36. [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
2. Boyd, S., Parikh, N., Chu, E., Peleato, B., & Eckstein, J. (2011). Distributed optimization and statistical learning via the alternating direction method of multipliers. Foundations and Trends in Machine Learning, 3(1), 1–122. [Crossref], [Google Scholar]
3. Fan, J., & Li, R. (2001). Variable selection via nonconcave penalized likelihood and its oracle properties. Journal of the American statistical Association, 96(456), 1348–1360. [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
4. Li, X., Zhou, Y., Zhu, Z., Liang, L., Yu, B., & Cao, W. (2018). Mapping annual urban dynamics (1985–2015) using time series of Landsat data. Remote Sensing of Environment, 216, 674–683. [Crossref], [Web of Science ®], [Google Scholar]
5. Ma, S., & Huang, J. (2017). A concave pairwise fusion approach to subgroup analysis. Journal of the American Statistical Association, 112(517), 410–423. [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
6. Ma, S., Huang, J., & Zhang, Z (2016). Exploration of heterogeneous treatment effects via concave fusion. arXiv preprint arXiv:1607.03717. [Google Scholar]
7. Molina, I., & Marhuenda, Y. (2015). sae: An R package for small area estimation. The R Journal, 7(1), 81–98. [Crossref], [Google Scholar]
8. Nusser, S. M., & Goebel, J. J. (1997). The national resources inventory: A long-term multi-resource monitoring programme. Environmental and Ecological Statistics, 4(3), 181–204. [Crossref], [Web of Science ®], [Google Scholar]
9. Ohio Supercomputer Center (2016). Owens supercomputer. http://osc.edu/ark:/19495/hpc6h5b1. [Google Scholar]
10. Pfeffermann, D. (2013). New important developments in small area estimation. Statistical Science, 28(1), 40–68. [Crossref], [Web of Science ®], [Google Scholar]
11. Rao, J. N., & Molina, I. (2015). Small area estimation. Hoboken, New Jersey: John Wiley & Sons. [Crossref], [Google Scholar]
12. Wang, H., Li, R., & Tsai, C.-L. (2007). Tuning parameter selectors for the smoothly clipped absolute deviation method. Biometrika, 94(3), 553–568. [Crossref], [Web of Science ®], [Google Scholar]
13. Wang, X., Zhu, Z., & Zhang, H. H (2019). Spatial automatic subgroup analysis for areal data with repeated measures. arXiv preprint arXiv:11906.01853. [Google Scholar]
14. Zhang, C.-H. (2010). Nearly unbiased variable selection under minimax concave penalty. The Annals of statistics, 38(2), 894–942. [Crossref], [Web of Science ®], [Google Scholar]
15. Zhu, R., Zou, G., Liang, H., & Zhu, L. (2016). Penalized weighted least squares to small area estimation. Scandinavian Journal of Statistics, 43(3), 736–756. [Crossref], [Web of Science ®], [Google Scholar]