Calibration Estimators in Survey Sampling

Deville, Jean-Claude; Särndal, Carl‐Erik

doi:10.2307/2290268

Cited by 542 publications

(777 citation statements)

References 0 publications

Supporting

Mentioning

757

Contrasting

Unclassified

Order By: Relevance

“…16 Overall characteristics of patients seeking access to each of the three types of GP and results reported here were based on weighted data. Distributions were compared using χ 2 and Fisher tests for categorical variables and Student and Wilcoxon tests for continuous variables.…”

Section: Resultsmentioning

confidence: 99%

Who seeks primary care for sleep, anxiety and depressive disorders from physicians prescribing homeopathic and other complementary medicine? Results from the EPI3 population survey

Grimaldi-Bensouda

Engel²,

Massol

et al. 2012

BMJ Open

View full text Add to dashboard Cite

ObjectivesTo describe and compare patients seeking treatment for sleep, anxiety and depressive disorders (SADD) from physicians in general practice (GPs) with three different practice preferences: strictly conventional medicine (GP-CM), mixed complementary and conventional medicine (GP-Mx) and certified homeopathic physicians (GP-Ho).Design and settingThe EPI3 survey was a nationwide, observational study of a representative sample of GPs and their patients, conducted in France between March 2007 and July 2008.Participants1572 patients diagnosed with SADD.Primary and secondary outcomesThe patients’ attitude towards complementary and alternative medicine; psychotropic drug utilisation.ResultsCompared to patients attending GP-CM, GP-Ho patients had healthier lifestyles while GP-Mx patients showed similar profiles. Psychotropic drugs were more likely to be prescribed by GP-CM (64%) than GP-Mx (55.4%) and GP-Ho (31.2%). The three groups of patients shared similar SADD severity.ConclusionOur results showed that patients with SADD, while differing principally in their sociodemographic profiles and conventional psychotropic prescriptions, were actually rather similar regarding the severity of SADD in terms of comorbidities and quality of life. This information may help to better plan resource allocation and management of these common health problems in primary care.

show abstract

Section: Resultsmentioning

confidence: 99%

Who seeks primary care for sleep, anxiety and depressive disorders from physicians prescribing homeopathic and other complementary medicine? Results from the EPI3 population survey

Grimaldi-Bensouda

Engel²,

Massol

et al. 2012

BMJ Open

View full text Add to dashboard Cite

show abstract

“…This has been performed mostly to improve the accuracy in stem volume and assortments (see Kangas and Maltamo 2000a;Kangas and Maltamo 2002;Mabvirura et al 2002). Note that the calibration estimation presented by Deville and Särndal (1992) necessitates abandoning the smooth shape of the original distribution function. Thus, calibration functions enable mimicking of the irregularities (e.g., bi-or multimodality) of the observed distributions (see Kangas and Maltamo 2000a;Kangas and Maltamo 2003).…”

Section: Discussionmentioning

confidence: 99%

Parameter recovery vs. parameter prediction for the Weibull distribution validated for Scots pine stands in Finland

Siipilehto¹,

Mehtätalo²

2013

Silva Fenn.

View full text Add to dashboard Cite

Highlights• A parameter recovery method (PRM) was developed for forest stand inventories and compared with previously developed parameter prediction methods (PPM) in Finland.• PRM for the 2-parameter Weibull function provided compatibility for the main stand characteristics: stem number, basal area and one of the four optional mean characteristics.• PRM provided comparable and at its best, superior accuracy in volume characteristics compared with PPM. AbstractThe moment-based parameter recovery method (PRM) has not been applied in Finland since the 1930s, even after a continuation of forest stand structure modelling in the 1980s. This paper presents a general overview of PRM and some useful applications. Applied PRM provided compatibility for the included stand characteristics of stem number (N) and basal area (G) with either mean (D), basal area-weighted mean (DG), median (DM) or basal area-median (DGM) diameter at breast height (dbh). A two-parameter Weibull function was used to describe the dbh-frequency distribution of Scots pine stands in Finland. In the validation, PRM was compared with existing parameter prediction models (PPMs). In addition, existing models for stand characteristics were used for the prediction of unknown characteristics. Validation consisted of examining the performance of the predicted distributions with respect to variation in stand density and accuracy of the localised distributions, as well as accuracy in terms of bias and the RMSE in stand characteristics in the independent test data set. The validation data consisted of 467 randomly selected stands from the National Forest Inventory based plots. PRM demonstrated excellent accuracy if G and N were both known. At its best, PRM provided accuracy that was superior to any existing model in Finland -especially in young stands (mean height < 9 m), where the RMSE in total and pulp wood volumes, 3.6 and 5.7%, respectively, was reduced by one-half of the values obtained using the best performing existing PPM (8.7-11.3%). The unweighted Weibull distribution solved by PRM was found to be competitive with weighted existing PPMs for advanced stands. Therefore, using PRM, the need for a basal area weighted distribution proved unnecessary, contrary to common belief. Models for G and N were shown to be unreliable and need to be improved to obtain more reliable distributions using PRM.

show abstract

“…One popular measure is case 2 from Deville and Sarndal [7],

\sum_{i, j} ρ_{i j} {g_{i j} \log (g_{i j}) - (g_{i j} - 1)}

, also recognizable as the deviance function from Poisson regression. Solving this optimization problem is equivalent to solving

\sum_{i j} \exp (- λ^{T} r_{i j}) ρ_{i j} r_{i j} = \sum_{i, j} r_{i j}

for λ.…”

Section: Estimating Equationmentioning

confidence: 99%

“…The consistency of the estimator

\overset{θ}{true}

readily follows from the fact that the estimating function (5) is asymptotically unbiased, which is partially due to the fact that

\hat{λ} = O_{p} true(1 ∕ \sqrt[]{N} true)

[7]. To find the asymptotic variance of the estimator, we first work out the asymptotic variance of U ( θ 0 ).…”

Section: Large Sample Theorymentioning

confidence: 99%

“…When there are auxiliary phase I covariates that are predictive of the missing phase II covariates, one naturally seeks to leverage that information to improve the estimation efficiency on the phase II covariate effects. A convergence of ideas from survey statisticians [7] and biostatisticians [8] showed that this could be done through adjustment of sampling weights.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Calibration weighted estimation of semiparametric transformation models for two‐phase sampling

Fong

Gilbert

2015

Statistics in Medicine

View full text Add to dashboard Cite

Two-phase designs are commonly used to sub-sample subjects from a cohort in order to study covariates that are too expensive to ascertain for everyone in the cohort. This is particularly true for the study of immune response biomarkers in vaccine immunology, where new, elaborate assays are constantly being developed to improve our understanding of the human immune responses to vaccines and how the immune response may protect humans from virus infection. It has long being recognized that if there exist variables that are correlated with expensive variables and can be measured for every subject in the cohort, they can be leveraged to improve the estimation efficiency for the effects of the expensive variables. In this paper, we develop an improved inverse probability weighted estimation approach for semiparametric transformation models with a two-phase study design. Semiparametric transformation models are a class of models that include the Cox proportional hazards and proportional odds models. They provide an attractive way to model the effects of immune response biomarkers as human immune responses generally wane over time. Our approach is based on weights calibration, which has its origin in survey statistics and was used by Breslow et al. [1,2] to improve inverse probability weighted estimation of the Cox regression model. We develop asymptotic theory for our estimator and examine its performance through simulation studies. We illustrate the proposed method with application to two HIV-1 vaccine efficacy trials.

show abstract

Calibration Estimators in Survey Sampling

Cited by 542 publications

References 0 publications

Who seeks primary care for sleep, anxiety and depressive disorders from physicians prescribing homeopathic and other complementary medicine? Results from the EPI3 population survey

Who seeks primary care for sleep, anxiety and depressive disorders from physicians prescribing homeopathic and other complementary medicine? Results from the EPI3 population survey

Parameter recovery vs. parameter prediction for the Weibull distribution validated for Scots pine stands in Finland

Calibration weighted estimation of semiparametric transformation models for two‐phase sampling

Contact Info

Product

Resources

About