African Journal of
Mathematics and Computer Science Research

  • Abbreviation: Afr. J. Math. Comput. Sci. Res.
  • Language: English
  • ISSN: 2006-9731
  • DOI: 10.5897/AJMCSR
  • Start Year: 2008
  • Published Articles: 261

Full Length Research Paper

Specification parameters for linear estimators in probability proportional to size sampling scheme

Ikughur, Atsua Jonathan
  • Ikughur, Atsua Jonathan
  • Department of Mathematics and Statistics Federal University Wukari, Taraba State, Nigeria
  • Google Scholar
Amahia, Godwin Nwanzu
  • Amahia, Godwin Nwanzu
  • Department of Statistics, University of Ibadan, Oyo State, Nigeria
  • Google Scholar


  •  Accepted: 27 February 2013
  •  Published: 30 April 2013

Abstract

 

Estimation of population parameters using the generalized moment estimators under probability proportional to size sampling scheme requires that the specification parameter, k defining these moments differs from one population to the other due to varying statistical properties of the study and measure of size variables. In this study, the approximate value of the specification parameter that minimizes the anticipated mean squared error is derived and is recommended for use in determining the best estimator for target populations in probability proportional to size (PPS) sampling especially, when the correlation between the study and auxiliary (measure of size) variables is being exploited. Empirical results also confirm that the specification parameter is a useful guide in defining appropriate estimator in PPS multi-character surveys.

 

Key words: Moment, estimation, specification parameter, transformation, correlation.