Articles |
Optimal sampling and estimation strategies under the linear model
Institute of Statistics, University of Neuchâtel, Pierre à Mazel 7, 2000 Neuchâtel, Switzerland desislava.nedyalkova{at}unine.ch yves.tille{at}unine.ch
Received for publication 1 February 2007. Revision received 1 January 2008.
In some cases model-based and model-assisted inferences can lead to very different estimators. These two paradigms are not so different if we search for an optimal strategy rather than just an optimal estimator, a strategy being a pair composed of a sampling design and an estimator. We show that, under a linear model, the optimal model-assisted strategy consists of a balanced sampling design with inclusion probabilities that are proportional to the standard deviations of the errors of the model and the Horvitz–Thompson estimator. If the heteroscedasticity of the model is 'fully explainable’ by the auxiliary variables, then this strategy is also optimal in a model-based sense. Moreover, under balanced sampling and with inclusion probabilities that are proportional to the standard deviation of the model, the best linear unbiased estimator and the Horvitz–Thompson estimator are equal. Finally, it is possible to construct a single estimator for both the design and model variance. The inference can thus be valid under the sampling design and under the model.
Key Words: Balanced sampling • Design-based inference • Finite population sampling • Fully explainable heteroscedasticity • Model-assisted inference • Model-based inference • Optimal strategy
References
-
Ardilly P. Échantillonnage représentatif optimum à probabilités inégales. Ann. DÉcon. Statist (1991) 23:91–113.
Brewer K. R. W. Survey sampling inference: Some past perspectives and present prospects. Pak. J. Statist (1994) 10:15–30.
Brewer K. R. W. Cosmetic calibration for unequal probability sample. Survey Methodol (1999a) 25:205–12.
Brewer K. R. W. Design-based or prediction-based inference? Stratified random vs stratified balanced sampling. Int. Statist. Rev (1999b) 67:35–47.[CrossRef]
Brewer K. R. W. Combined Survey Sampling Inference, Weighing Basu's Elephants (2002) London: Arnold.
Brewer K. R. W., Hanif M., Tam S. M. How nearly can model-based prediction and design-based estimation be reconciled. J. Am. Statist. Assoc (1988) 83:128–32.[CrossRef][Web of Science]
Chambers R. L. Robust case-weighting for multipurpose establishment surveys. J. Offic. Statist (1996) 12:3–32.
Cumberland W. G., Royall R. M. Prediction models in unequal probability sampling. J. R. Statist. Soc (1981) B 43:353–67.
Deville J.-C. Constrained samples, conditional inference, weighting: Three aspects of the utilisation of auxiliary information. Proc. Workshop on the Uses of Auxiliary Information in Surveys (1992) Örebro, Sweden: Statistics Sweden. 21–40.
Deville J.-C. Variance estimation for complex statistics and estimators: Linearization and residual techniques. Survey Methodol (1999) 25:193–204.
Deville J.-C., Grosbras J.-M., Roth N. Efficient sampling algorithms and balanced samples. In: COMPSTAT, Proceedings in Computational Statistics—Payne R., Green P., eds. (1988) Heidelberg: Physica. 255–66.
Deville J.-C., Särndal C.-E. Calibration estimators in survey sampling. J. Am. Statist. Assoc (1992) 87:376–82.[CrossRef][Web of Science]
Deville J.-C., Tillé Y. Efficient balanced sampling: The cube method. Biometrika (2004) 91:893–912.
Deville J.-C., Tillé Y. Variance approximation under balanced sampling. J. Statist. Plan. Infer (2005) 128:411–25.[CrossRef]
Godambe V. P. A unified theory of sampling from finite population. J. R. Statist. Soc (1955) B 17:269–78.
Godambe V. P., Joshi V. M. Admissibility and Bayes estimation in sampling finite populations I. Ann. Math. Statist (1965) 36:1707–22.[CrossRef]
Hájek J. Sampling from a Finite Population (1981) New York: Marcel Dekker.
Hansen M. H., Madow W. G., Tepping B. J. An evaluation of model dependent and probability-sampling inferences in sample surveys (with Discussion). J. Am. Statist. Assoc (1983) 78:776–807.[CrossRef][Web of Science]
Hedayat A. S., Majumdar D. Generating desirable sampling plans by the technique of trade off in experimental design. J. Statist. Plan. Infer (1995) 44:237–47.[CrossRef]
Iachan R. Sampling strategies, robustness and efficiency: The state of the art. Int. Statist. Rev (1984) 52:209–18.
Isaki C. T., Fuller W. A. Survey design under a regression population model. J. Am. Statist. Assoc (1982) 77:89–96.[CrossRef][Web of Science]
Kott P. S. When a mean-of-ratios is the best linear unbiased estimator under a model. Am. Statist (1986) 40:202–4.[CrossRef]
Royall R. M. The linear least squares prediction approach to two-stage sampling. J. Am. Statist. Assoc (1976) 71:657–64.[CrossRef][Web of Science]
Royall R. M. The prediction approach to sampling theory. Handbook of Statistics (1988) 6. Amsterdam, Holland: Elsevier Science Publishers. 399–413.[CrossRef]
Royall R. M. Robustness and optimal design under prediction models for finite populations. Survey Methodol (1992) 18:179–85.
Royall R. M., Cumberland W. G. The finite population linear regression estimator and estimators of its variance. An empirical study. J. Am. Statist. Assoc (1981) 76:924–30.[CrossRef][Web of Science]
Royall R. M., Herson J. Robust estimation in finite populations I. J. Am. Statist. Assoc (1973a) 68:880–9.[CrossRef][Web of Science]
Royall R. M., Herson J. Robust estimation in finite populations II: Stratification on a size variable. J. Am. Statist. Assoc (1973b) 68:891–3.
Särndal C.-E., Swensson B., Wretman J. H. Model Assisted Survey Sampling (1992) New York: Spinger.
Scott A. J., Brewer K. R. W., Ho E. W. H. Finite population sampling and robust estimation. J. Am. Statist. Assoc (1978) 73:359–61.[CrossRef][Web of Science]
Smith T. M. F. The foundations of survey sampling: A review. J. R. Statist. Soc (1976) A 139:183–204.
Smith T. M. F. Sample surveys, present position and potential developments: Some personal views (with Discussion). J. R. Statist. Soc (1984) A 147:208–21.
Smith T. M. F. Sample surveys 1975–1990; An age of reconciliation (with Discussion). Int. Statist. Rev (1994) 62:5–34.
Thionet P. La théorie des sondages (1953) Paris: INSEE, Imprimerie Nationale.
Tillé Y. Sampling Algorithms (2006) New York: Springer.
Valliant R., Dorfman A. H., Royall R. M. Finite Population Sampling and Inference: A Prediction Approach (2000) New York: Wiley.
Yates F. A review of recent statistical developments in sampling and sampling surveys (with Discussion). J. R. Statist. Soc (1946) A 109:12–43.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||