This paper fills the gap in the proof of asymptotic normality of weighted likelihood estimators for parameters fitted to two-phase stratified samples when sampling weights were estimated by the data in the authors’ paper ibid. 34, No. 1, 86--102 (2007; Zbl 1142.62014
). The proof is corrected under slightly strengthened assumptions on the theorem. A new theorem on the asymptotic behavior of estimating equations estimates in presence of infinite dimensional nuisance parameters is used to complete the proof.
Reviewer: R. E. Maiboroda (Kyïv)