Showing 1 - 10 of 52,096
In longitudinal studies, analysis can be based on any one of a large number of wavecombinations. However, only one set of non-response weights (often based on respondents from all waves up to the latest) is typically offered on public use data files. We refer to this as a single weighting...
Persistent link: https://www.econbiz.de/10010422757
Persistent link: https://www.econbiz.de/10011312177
Persistent link: https://www.econbiz.de/10011691119
Persistent link: https://www.econbiz.de/10012174823
less data. The choice of survey mode, therefore, involves a potential tradeoff between bias and variance of estimators. I …
Persistent link: https://www.econbiz.de/10012425355
incomplete sample. The proposed strategies were found to be more efficient and bias-controllable than similar estimation …
Persistent link: https://www.econbiz.de/10014287900
Persistent link: https://www.econbiz.de/10014423488
Persistent link: https://www.econbiz.de/10001508154
Persistent link: https://www.econbiz.de/10000893765
second normal distribution of low weight. In both cases of misspecification the bias, of course, tends to zero when the size … of misspecification tends to zero. However, in the first case the bias goes to zero in a flat way so that small … deviations from the true model lead to a negligible bias, whereas in the second case the bias is noticeable even for small …
Persistent link: https://www.econbiz.de/10002719769