Design-based estimators of totals, means or proportions in finite populations generally are functions of weighted sums. If each element selected into the sample is also observed, then for the calculation of the pi-estimator these weights are just the inverse inclusion probabilities of the elements. However, if e.g. nonresponse or attrition over time occurs, calculation of these weights also includes ...
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 ...
A simulation study designed to evaluate the pseudo-R2 T proposed by Spiess and Keller (1999) suggests that this measure represents the goodness- of-fit not only of the systematic part, but also of the assumed correlation structure in binary panel probit models.
The results of two simulation studies suggest a mixed 'generalized estimating/pseudo-score equations' approach to lead to more efficient estimators than a GEE approach proposed by Qu, Williams, Beck and Medendorp (1992) or a three-stage approach as proposed e.g. by Schepers, Arminger and Küsters (1991) in panel probit models with binary responses. Furthermore, the mixed approach led to very efficient ...
In:
Jürgen Chlumsky, Bernhard Schimpl-Neimanns, Gert G. Wagner (Hrsg.) ,
Kooperation zwischen Wissenschaft und amtlicher Statistik
Stuttgart : Metzler-Poeschel
S. 59-61
Forum der Bundesstatistik : Schriftenreihe ; 34
| Gert G. Wagner