On some modification of the sum-quota sampling scheme
Languages of publication
A simple extension of the well-known sequential sum-quota sampling scheme is proposed. The modification facilitates the simultaneous examination of several sampling units between checks of the cost limit. This may speed up the data gathering process while some degree of control over the variable sample cost is still retained. It is proposed to estimate the population total of the studied characteristic under such a sampling scheme using empirical estimates of inclusion probabilities evaluated in a simulation study. It appears that at least in some situations the empirical Horvitz-Thompson estimator is approximately unbiased.
- Fattorini L. (2006). Applying the Horvitz-Thompson criterion in complex designs: A computer-intensive perspective for estimating inclusion probabilities. Biometrika. Vol. 93(2). Pp. 269-278.
- Groves R.M. (1989). Survey Errors and Survey Costs. John Wiley & Sons. New York.
- Kremers W.K. (1985). The Statistical Analysis of Sum-Quota Sampling. Unpublished PHD thesis. Cornell University.
- Kremers W.K. (1986). Completeness and unbiased estimation for sum-quota sampling. Journal of the American Statistical Association. Vol. 81(396). Pp. 1070-1073.
- Kremers W.K., Robson D. (1987). Unbiased estimation when sampling from renewal processes: The single sample and k-sample random means cases. Biometrika. Vol. 74(2). Pp. 329-336.
- Lehmann E.L., Casella G. (1998). Theory of Point Estimation. Springer. New York.
- Pathak K. (1976). Unbiased estimation in fixed-cost sequential sampling schemes. Annals of Statistics. Vol. 4(5). Pp. 1012-1017.
Publication order reference