A Class of Dual Frame Survey Sampling Estimators in the Presence of a Covariate: How Amy Predicts Her President

• Activity Number: 654
• Type: Contributed
• Date/Time: Thursday, August 7, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Survey Research Methods Section
• Author(s): Sarjinder Singh*+ and Stephen Andrew Sedory and David Molina
• Companies: Texas A&M and Texas A&M and University of Granada
• Keywords: Dual frame survey ; Estimation of population total ; Power transformation ; Co-variate

Abstract

In this paper, a fictitious story, "How Amy Predicts Her President", is introduced to motivate the research considered. In the course of the story we propose a new class of estimators in dual frame survey sampling that makes use of a power transformation. The estimator proposed by Hartley (1962, 1974) is shown to be a special case of the proposed class of estimators. The mean squared error of the proposed estimator is derived and compared to that of the Hartley estimator. A suggestion is given for improving the Fuller and Burmeister (1972) estimator along similar lines. Lastly, the work is extend to the case of multi-covariates. Note that we make no use of any known parameter of auxiliary information as in the ratio estimator due to Cochran (1940). In this regard the proposed class of estimators is different from the existing estimators in the literature of dual frame survey sampling. We show theoretically that the proposed class of estimators is always more efficient than the pioneer Hartley (1962, 1974) estimator. The results are also justified through extensive simulated numerical situations