Prediction of ordered random effects in a simple small area model
Rinott, Yosef and Malinovsky, Yaakov (2008) Prediction of ordered random effects in a simple small area model. [Discussion Paper]. p. 20. (Submitted)
|PDF (Full text) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader|
Official URL: http://pluto.huji.ac.il/~rinott/publications/Yakov...
Prediction of a vector of ordered parameters or part of it arises naturally in the context of Small Area Estimation (SAE). For example, one may want to estimate the parameters associated with the top ten areas, the best or worst area, or a certain percentile. We use a simple SAE model to show that estimation of ordered parameters by the corresponding ordered estimates of each area separately does not yield good results with respect to MSE. Shrinkage-type predictors, with an appropriate amount of shrinkage for the particular problem of ordered parame- ters, are considerably better, and their performance is close to that of the optimal predictors, which cannot in general be computed explicitly.
|Item Type:||Report / Paper (Discussion Paper)|
|Research documents and activity classification:||Working Papers > Non-Refereed Working Papers / of national relevance only|
|Divisions:||Department of Business and Management|
|Additional Information:||Preprint (2008) to appear in "Statistica Sinica".|
|Uncontrolled Keywords:||Empirical Bayes predictor. Shrinkage. Order statistics. Linear predictor.|
|MIUR Scientific Area:||Area 13 - Economics and Statistics > SECS-S/01 Statistics|
|Deposited By:||Chiara Annulli (admin)|
|Deposited On:||28 Oct 2009 17:30|
|Last Modified:||17 Apr 2013 20:00|
Repository Staff Only: item control page