1. Division of Mathematics Computer Science and Statistics , University of Texas at San Antonio , San Antonio, 78285, Texas;2. Department of Mathematics and Statistics , University of New Mexcico , Albuquerque, 87131, New Mexico
Abstract:
In the case of a finite population with a linear trend, improvements in estimation procedures are suggested for several well-known systematic samplingschemes. The use of the least squares regression estimator is examined and a sampling scheme which reduces its expected mean squared error is proposed.