Multiply robust imputation procedures for zero-inflated distributions in surveys

Chen, Sixia; Haziza, David

doi:10.1007/s40300-017-0128-9

Cited by 35 publications

(63 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although doubly robust methods offer some protection, point and variance estimators may perform poorly in terms of bias and efficiency if both models are misspecified. This point was raised by a number of authors, including Kang and Schafer (2007) and Chen and Haziza (2017).…”

Section: Introductionmentioning

confidence: 99%

“…To provide additional protection, Han and Wang (2013) introduced the concept of multiple robustness; see also Han (2014aHan ( , 2014bHan ( , 2016aHan ( , 2016b, Chan and Yam (2014), Chen and Haziza (2017) and Duan and Yin (2017).…”

Section: Introductionmentioning

confidence: 99%

“…An estimation procedure is said to be multiply robust if it remains consistent if any one of these multiple models is correctly specified. Multiply robust procedures are attractive in many practical situations; see Chen and Haziza (2017) for a discussion. When all the models are misspecified, a number of empirical investigations available in the literature suggest that multiply robust estimation procedures perform much better than doubly robust procedures based on the same working models, in terms of bias and efficiency; see e.g., Han (2014b) and Chen and Haziza (2017).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Multiply robust nonparametric multiple imputation for the treatment of missing data

Chen¹,

Haziza²

2019

STAT SINICA

Self Cite

View full text Add to dashboard Cite

Imputation is an effective way for handling missing values. We propose a nonparametric multiple imputation procedure that makes use of multiple outcome regression models and/or multiple propensity score models. Our procedure leads to a multiply robust point estimator in the sense that it remains consistent if all the models but one are misspecified. A variance estimate is readily obtained by applying the customary rule advocated by Rubin (1987). The asymptotic properties of the proposed method are established. Results from a simulation study, assessing the proposed method in terms of bias, efficiency and coverage probability, support our findings.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Multiply robust nonparametric multiple imputation for the treatment of missing data

Chen¹,

Haziza²

2019

STAT SINICA

Self Cite

View full text Add to dashboard Cite

show abstract

“…that target informativeness ( [2,6,7]) and nonsampling errors such as nonresponse ( [4,6]). All contributions account for the complexity of the sampling design and make use of models.…”

mentioning

confidence: 99%

“…Nonresponse is addressed by Haziza and Chen [4] as well. Differently from [6], item non-response is considered and treated by means of multiple imputation.…”

mentioning

confidence: 99%

Foreword to the special issue on “Advances in Survey Statistics”

Opsomer

Ranalli

Dickson

2017

METRON

View full text Add to dashboard Cite

This Special Issue was inspired by the desire to bring together valuable and modern contributions in the field of Survey Statistics, which shed light on some of the challenging tasks of designing and analyzing surveys. Traditionally, survey statistics has been seen as a separate branch to the rest of statistics, because of its reliance on the design-based approach to inference, as opposed to the model-based approach used in other applications of statistics. The former approach is certainly sensible for survey data, because it is common for design features such as strata, clusters and unequal probabilities to induce strong biasing effects in the data. While these effects can be incorporated in models, an often simpler and more robust approach is to rely on randomization theory and inverse-probability weighting to produce estimates and perform inference.However, there is also a need to recognize the many interactions between randomizationbased inference and statistical modeling. The role of modeling in regression estimation is well understood and has been incorporated into the design-based framework for several decades. At the other extreme, small area estimation methods are fully model-based, with the design effects incorporated through the model or, when appropriate, ignored. But there are many statistical research topics that occur at the intersection of randomization-based and model-based inference, as illustrated by many of the contributions to this Special Issue. These include nonresponse and measurement errors in survey data, estimation under non-measurable designs, and fitting of statistical models under design informativeness. The contributions of this Special Issue deal with topics in either design ([1,7]) or estimation ([2-6]), also including approaches B M. Giovanna Ranalli

show abstract

Pseudo empirical likelihood inference for nonprobability survey samples

Chen

Rao

et al. 2022

Can J Statistics

View full text Add to dashboard Cite

In this article, we first provide an overview of two major developments on complex survey data analysis: the empirical likelihood methods and statistical inference with nonprobability survey samples. We highlight the important research contributions to the field of survey sampling in general and the two topics in particular by Canadian survey statisticians. We then propose new inferential procedures for analyzing nonprobability survey samples through the pseudo empirical likelihood approach. The proposed methods lead to point estimators asymptotically equivalent to those discussed in the recent literature but with more desirable features on confidence intervals such as range-respecting and data-driven orientation. Results from a simulation study demonstrate the superiority of the proposed methods in dealing with binary response variables.Résumé: Les auteurs de ce travail présentent, dans un premier temps, un aperçu de deux développements majeurs en analyse de données d'enquête complexes, à savoir les méthodes de vraisemblance empirique et l'inférence statistique en sondages non probabilistes. Ensuite, ils mettent en évidence les importantes contributions de statisticiens canadiens à la recherche en enquêtes par sondage en général et à ces deux sujets en particulier. En faisant appel à l'approche de pseudo-vraisemblance empirique, ils proposent de nouvelles procédures d'inférence pour l'analyse d'échantillons d'enquêtes non probabilistes. Ces méthodes conduisent à des estimateurs ponctuels asymptotiquement équivalents à des estimateurs discutés dans les écrits récents mais possédant des intervalles de confiance qui jouissent de meilleures propriétés, telles que préserver l'étendue et être axés sur les données. Les résultats d'une étude de simulation démontrent la supériorité des méthodes proposées dans le traitement de variables de réponse binaire.

show abstract

Multiply robust imputation procedures for zero-inflated distributions in surveys

Cited by 35 publications

References 26 publications

Multiply robust nonparametric multiple imputation for the treatment of missing data

Multiply robust nonparametric multiple imputation for the treatment of missing data

Foreword to the special issue on “Advances in Survey Statistics”

Pseudo empirical likelihood inference for nonprobability survey samples

Contact Info

Product

Resources

About