Please use this identifier to cite or link to this item: http://hdl.handle.net/10419/18194
Authors: 
Spiess, Martin
Rendtel, Ulrich
Year of Publication: 
2000
Series/Report no.: 
DIW Discussion Papers 198
Abstract: 
In this paper, a weight is derived for the calculation of design based estimators of totals, means and proportions using the ongoing German Socio-Economic Panel (GSOEP) and a new cross-sectional sample. In the first part of the paper, the selection schemes of the subsamples A, B, C and D of the ongoing panel and of the new sample (sample E) are described. Using some similarity assumptions and starting from a general formulation, an optimal weight in the sense of small variances of design-based estimators using both samples is derived. The merits of this approach as well as some disadvantages are discussed.
Subjects: 
design based inference
convex weighting estimator
complex surveys
panel survey
cross-sectional sample
Document Type: 
Working Paper

Files in This Item:
File
Size
214.5 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.