Calibration Estimators in Survey Sampling

By J. Deville, C. Särndal. 1992

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Nonparametric Model Calibration Estimation in Survey Sampling (2005, methodology): “More general situations are handled by generalized regression estimation (Särndal 1980; Särndal, Swensson, and Wretman 1992) and calibration estimation (Deville and Särndal 1992).”
Multiply robust imputation procedures for the treatment of item nonresponse in surveys (2017, background): “Furthermore, the function satisfies G(w̃i/wi) 0 and G(1) = 0; see Deville & Särndal (1992).”
Spatial prediction of forest stand variables (2009, methodology): “Deville and Särndal (1992) present regression estimators with weights for the auxiliary variables being calibrated, so that weights are as close to the design-based weights as possible.”
Estimation With Survey Data Under Nonignorable Nonresponse or Informative Sampling (2002, methodology): “By using calibration, Deville and Särndal (1992) investigated estimation of nite population totals in the presence of known population totals of covariates.”
Using Calibration Weighting to Adjust for Nonignorable Unit Nonresponse (2010, unknown): “For example, Deville and Särndal (1992) have shown that if the wk are chosen to minimize ∑ k∈S(wk −dk)2/dk such that Tx = ∑ k∈S wkxk, then wk = dk(1 + xTk g), where g = (∑j∈S djxjxTj )−1(Tx − t̂xπ ).”
Teacher Attrition and Mobility: Results from the 2012-13 Teacher Follow-Up Survey. First Look. NCES 2014-077. (2014, methodology): “This improves the precision of survey estimates (Deville and Särndal 1992).”
Calibration estimation using empirical likelihood in survey sampling (2009, methodology): “The classical regression estimator described in Cochran (1977) uses a Euclidian distance function.”
Globally efficient non-parametric inference of average treatment effects by empirical balancing calibration weighting (2015, methodology): “The general idea of calibration as in Deville and Särndal (1992) is to minimize the aggregate distance between the final weights w = .w1, : : : , wN/ to a given vector of design weights d = .d1, : : : , dN/ subject to moment constraints.”
Variance estimation for complex indicators of poverty and inequality using linearization techniques (2009, methodology): “When sample weights are adjusted to auxiliary variables which are correlated with the survey target variables, one can expect the accuracy to be better (Deville and Särndal, 1992).”
Calibration Estimation in Survey Sampling (2010, unknown): “Instead of using (19), Deville and S¨arndal (1992) proposed a variance estimator of the form (20) with If h i ( λ ) satisﬁes (16), the term (23) is equivalent to (19).”
Modelling the diameter distribution of Pinus sylvestris, Pinus nigra and Pinus halepensis forest stands in Catalonia using the truncated Weibull function (2006, methodology): “However, since the predicted distribution may sometimes substantially deviate from the true distribution, calibration estimation could be used to improve the fi t (see Deville and Särndal, 1992 ; Kangas and Maltamo, 2003 ).”
Covariate balancing propensity score by tailored loss functions (2016, background): “They belong to a general distance measure ∑n i=1D(wi, vi) in Deville and Särndal [11], where D(w,v) is a continuously differentiable and strongly convex function in w and achieves its minimum at w = v.”
Taylor linearization sampling errors and design effects for poverty measures and other complex statistics (2011, background): “The impact of calibration on variance can be highly dependent on the particular situation, and special procedures are required to take this effect into account [10].”