2 p.m. - 3 p.m. Location: FN 2.102
Prof. Lynne Stokes
Dept of Statistical Science
Southern Methoodist University
Measurement Error in Dual Frame Designs
Estimators of population parameters made from for surveys that use more than one overlapping frame for collecting data require knowledge of frame membership for each selected unit. In some cases this information can be obtained directly from the frames. In other cases, it can only be obtained from the responding unit itself. In either case, it is possible that the information about frame membership is erroneous. The effect of this measurement error is bias in estimators of means and totals for the population. In some cases, it is possible to collect precise data on a subset of the sample. In this paper, we propose a new estimator that makes use of this double sample for mitigating the bias. The new estimator is used for estimating catch from a dual frame survey of recreational marine anglers.
Refreshments will be served in the alcove outside of FN 2.102 30 minutes prior to the talk
Sponsored by the Department of Mathematical Sciences
John Zweck, 972-883-6699
Questions? Email me.