Page 210 - Contributed Paper Session (CPS) - Volume 2

CPS1824 Sanggi L.

Figure-4: Household split Figure-5: Household merge

4. Conclusion
The wave 1 selection probabilities of new entrants are generally unknown
in household panel surveys. To solve this problem the weight sharing method
and the modeling method have been implemented in most panel surveys. But
in the administrative data-based KPCLS, the histories of the household and the
individual are managed annually so that the selection probability for the
current household can be calculated directly which is theoretically and
empirically more accurate method. And this method makes it possible to make
the various cross-sectional and longitudinal weighting schemes simpler. As
using administrative data became more important, we hope that this paper
will be a good reference in other administrative panels in the future.

References
1. Schonlau. M., Kroh. M. and Watson. N. (2013). The implementation of
cross-sectional weights in household panel surveys. Statistics Surveys,
Vol. 7, 37-57.
2. Watson. N. (2012). Longitudinal and Cross-sectional Weighting
Methodology for the HILDA Survey. HILDA Project technical paper
series.

199 | I S I W S C 2 0 1 9