Doubly Robust Regression Analysis for Data Fusion

Evans, Katherine; Sun, Baoluo; Robins, James M.; Tchetgen, Eric J. Tchetgen

doi:10.5705/ss.202018.0334

Cited by 4 publications

(15 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Unfortunately, the CEX provides limited information on wealth data and thus the CEX alone is not sufficient for our purposes. Previous research (Bostic et al, 2009) (Bostic et al, 2009), we truncate the SCF sample at 90th percentiles of observed total household income and net worth, as was done in Evans et al (2021).…”

Section: Applicationmentioning

confidence: 99%

“…Because there are missing values in the original survey data, the publicly available data set consists of five imputed replicates. We thus follow Evans et al (2021) to perform estimation for each replicate and combine the results using Rubin's rule (Rubin, 2004). We report the point estimates and standard errors of our analysis results using all the four working models in Table 2.…”

Section: Applicationmentioning

confidence: 99%

“…Recently, Graham et al (2016) proposed a more flexibly parametric modeling approach, which is doubly robust in that it is consistent if the propensity score or the imputation model is correctly specified in a certain class of nuisance models. To break this limitation, Shu and Tan (2020) and Evans et al (2021) developed general doubly robust estimators with unrestricted nuisance model specifications. Doubly robust procedures offer some protection against model misspecification, but as only a single propensity score model and a single imputation model are allowed, they may not provide sufficient protection and it is restrictive to assume that either model is correctly specified in practice.…”

Section: Introductionmentioning

confidence: 99%

“…Under some scenarios, the global identifiability of θ 0 can be guaranteed. For example, when µ(•) is the identity function, θ 0 is identifiable if there exists a nonlinear term of V in E(W | V ) (Evans et al, 2021;Miao et al, 2021). In the next section, we assume θ 0 has been identified and propose a procedure to estimate θ 0 based on the moment condition (1) with a fixed function t(V ; θ).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Calibrated regression estimation using empirical likelihood under data fusion

Li¹,

Luo²,

Xu³

2022

Preprint

View full text Add to dashboard Cite

Data analysis based on information from several sources is common in economic and biomedical studies. This setting is often referred to as the data fusion problem, which differs from traditional missing data problems since no complete data is observed for any subject. We consider a regression analysis when the outcome variable and some covariates are collected from two different sources. By leveraging the common variables observed in both data sets, doubly robust estimation procedures are proposed in the literature to protect against possible model misspecifications. However, they employ only a single propensity score model for the data fusion process and a single imputation model for the covariates available in one data set. It may be questionable to assume that either model is correctly specified in practice. We therefore propose an approach that calibrates multiple propensity score and imputation models to gain more protection based on empirical likelihood methods. The resulting estimator is consistent when any one of those models is correctly specified and is robust against extreme values of the fitted propensity scores. We also establish its asymptotic normality property and discuss the semiparametric estimation efficiency. Simulation studies show that the proposed estimator has substantial advantages over existing doubly robust estimators, and an assembled U.S. household expenditure data example is used for illustration.

show abstract

Section: Applicationmentioning

confidence: 99%

Section: Applicationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Calibrated regression estimation using empirical likelihood under data fusion

Li¹,

Luo²,

Xu³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…We consider a regression model for the outcome on all available covariates measured in the internal study, while using information from multiple external data sources, which Evans et al (2018) refer to as the regression analysis for data fusion.…”

Section: Introductionmentioning

confidence: 99%

A synthetic data integration framework to leverage external summary-level information from heterogeneous populations

Gu¹,

Taylor²,

Mukherjee³

2021

Preprint

View full text Add to dashboard Cite

There is a growing need for flexible general frameworks that integrate individual-level data with external summary information for improved statistical inference. External information relevant for a risk prediction model may come in multiple forms, through regression coefficient estimates or predicted values of the outcome models.Different external models may use different sets of predictors and the algorithm they used to predict the outcome Y given these variables may or may not be known. The underlying populations corresponding to each external model may be different and they may differ from the internal study population. This paper proposes an imputation-based methodology where the goal is to fit an outcome regression model with all available variables in the internal study while utilizing summary information from external models that may have used only a subset of the predictors. The method allows for heterogeneity of covariate effects across the external populations. The proposed approach generates synthetic outcome data in each population, uses stacked multiple imputation to create a long dataset with complete covariate information, and finally analyzes the imputed data with weighted regression. This flexible and unified approach attains the following four objectives: (i) incorporating supplementary information from a broad class of externally fitted predictive models or established risk calculators which could be based on parametric regression or machine learning methods, as long as the external model can generate outcome values given covariates; (ii) improving statistical efficiency of the estimated coefficients in the internal study; (iii) improving predictions by utilizing even partial information available from models that uses a subset of the full set of covariates used in the internal study; and (iv) providing valid statistical inference for the external population with potentially different covariate effects from the internal population. Applications include prostate cancer risk prediction models using novel biomarkers that are measured only in the internal study.

show abstract

Mediation analysis using incomplete information from publicly available data sources

Derkach,

Kantor,

Sampson

et al. 2024

Statistics in Medicine

View full text Add to dashboard Cite

Our work was motivated by the question whether, and to what extent, well‐established risk factors mediate the racial disparity observed for colorectal cancer (CRC) incidence in the United States. Mediation analysis examines the relationships between an exposure, a mediator and an outcome. All available methods require access to a single complete data set with these three variables. However, because population‐based studies usually include few non‐White participants, these approaches have limited utility in answering our motivating question. Recently, we developed novel methods to integrate several data sets with incomplete information for mediation analysis. These methods have two limitations: (i) they only consider a single mediator and (ii) they require a data set containing individual‐level data on the mediator and exposure (and possibly confounders) obtained by independent and identically distributed sampling from the target population. Here, we propose a new method for mediation analysis with several different data sets that accommodates complex survey and registry data, and allows for multiple mediators. The proposed approach yields unbiased causal effects estimates and confidence intervals with nominal coverage in simulations. We apply our method to data from U.S. cancer registries, a U.S.‐population‐representative survey and summary level odds‐ratio estimates, to rigorously evaluate what proportion of the difference in CRC risk between non‐Hispanic Whites and Blacks is mediated by three potentially modifiable risk factors (CRC screening history, body mass index, and regular aspirin use).

show abstract

Doubly Robust Regression Analysis for Data Fusion

Cited by 4 publications

References 29 publications

Calibrated regression estimation using empirical likelihood under data fusion

Calibrated regression estimation using empirical likelihood under data fusion

A synthetic data integration framework to leverage external summary-level information from heterogeneous populations

Mediation analysis using incomplete information from publicly available data sources

Contact Info

Product

Resources

About