What question did this study set out to answer?

To develop a general framework for survey calibration estimators that derive sample weighting systems by minimizing distance functions subject to calibration constraints on auxiliary variables.

June 1, 1992Journal of the American Statistical Association

Calibration Estimators in Survey Sampling

View Full Paper

Ask AI

Bookmark

Authors

JDJean‐Claude DevilleUniversité Rennes 2 CSCarl‐Erik SärndalStatistics Sweden

Discussion

Member takes

Implication

Methodological study demonstrates a unified calibration framework for survey weights, indicating asymptotic equivalence to general regression estimators.

Key Points

To develop a general framework for survey calibration estimators that derive sample weighting systems by minimizing distance functions subject to calibration constraints on auxiliary variables.
Formulated a constrained optimization problem minimizing distance measures between calibrated weights and design inclusion weights subject to calibration equations.
Evaluated alternative distance functions to resolve practical limitations of the general regression estimator, such as negative weights.
Applied the framework to contingency tables with known marginal counts through generalized raking procedures.
Demonstrated that the general regression estimator represents a first-order approximation to all calibration estimators within the proposed family.
Established asymptotic equivalence between all family members and the general regression estimator, justifying the use of standard GREG variance estimation across all distance measures.
Showed that practical estimator selection can rely primarily on computational tractability and desired numerical properties of the weights, including non-negativity.

Cite This Study

Deville et al. (1992) studied this question.

synapsesocial.com/papers/6a7ec4da14ec6a5b094f7bd1 https://doi.org/10.2307/2290268

View Full Paper

Ask AI

Bookmark

Implication

Methodological study demonstrates a unified calibration framework for survey weights, indicating asymptotic equivalence to general regression estimators.

Calibration Estimators in Survey Sampling

Discussion

Member takes

Implication

Key Points

Cite This Study

Calibration Estimators in Survey Sampling

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider