Ensemble methods using the same underlying algorithm trained on different subsets of observations have recently received increased attention as practical prediction tools for massive datasets. We propose Subsemble: a general subset ensemble prediction method, which can be used for small, moderate, or large datasets. Subsemble partitions the full dataset into subsets of observations, fits a specified underlying algorithm on each subset, and uses a clever form of V-fold cross-validation to output a prediction function that combines the subset-specific fits. We give an oracle result that provides a theoretical performance guarantee for Subsemble. Through simulations, we demonstrate that Subsemble can be a beneficial tool for small to moderate sized datasets, and often has better prediction performance than the underlying algorithm fit just once on the full dataset. We also describe how to include Subsemble as a candidate in a SuperLearner library, providing a practical way to evaluate the performance of Subsemlbe relative to the underlying algorithm fit just once on the full dataset.
doi:10.1080/02664763.2013.864263
PMCID: PMC4000126
PMID: 24778462
ensemble methods; prediction; cross-validation; machine learning; big data
The cross-validation deletion–substitution–addition (cvDSA) algorithm is based on data-adaptive estimation methodology to select and estimate marginal structural models (MSMs) for point treatment studies as well as models for conditional means where the outcome is continuous or binary. The algorithm builds and selects models based on user-defined criteria for model selection, and utilizes a loss function-based estimation procedure to distinguish between different model fits. In addition, the algorithm selects models based on cross-validation methodology to avoid “over-fitting” data. The cvDSA routine is an R software package available for download. An alternative R-package (DSA) based on the same principles as the cvDSA routine (i.e., cross-validation, loss function), but one that is faster and with additional refinements for selection and estimation of conditional means, is also available for download. Analyses of real and simulated data were conducted to demonstrate the use of these algorithms, and to compare MSMs where the causal effects were assumed (i.e., investigator-defined), with MSMs selected by the cvDSA. The package was used also to select models for the nuisance parameter (treatment) model to estimate the MSM parameters with inverse-probability of treatment weight (IPTW) estimation. Other estimation procedures (i.e., G-computation and double robust IPTW) are available also with the package.
doi:10.1016/j.csda.2010.02.002
PMCID: PMC4259156
PMID: 25505354
Cross-validation; Machine learning; Marginal structural models; Lung function; Cardiovascular mortality
The PROmotion of Breastfeeding Intervention Trial (PROBIT) cluster-randomized a program encouraging breastfeeding to new mothers in hospital centers. The original studies indicated that this intervention successfully increased duration of breastfeeding and lowered rates of gastrointestinal tract infections in newborns. Additional scientific and popular interest lies in determining the causal effect of longer breastfeeding on gastrointestinal infection. In this study, we estimate the expected infection count under various lengths of breastfeeding in order to estimate the effect of breastfeeding duration on infection. Due to the presence of baseline and time-dependent confounding, specialized “causal” estimation methods are required. We demonstrate the double-robust method of Targeted Maximum Likelihood Estimation (TMLE) in the context of this application and review some related methods and the adjustments required to account for clustering. We compare TMLE (implemented both parametrically and using a data-adaptive algorithm) to other causal methods for this example. In addition, we conduct a simulation study to determine (1) the effectiveness of controlling for clustering indicators when cluster-specific confounders are unmeasured and (2) the importance of using data-adaptive TMLE.
PMCID: PMC4259272
PMID: 25505499
Causal inference; G-computation; inverse probability weighting; marginal effects; missing data; pediatrics
Due to the need to evaluate the effectiveness of community-based programs in practice, there is substantial interest in methods to estimate the causal effects of community-level treatments or exposures on individual level outcomes. The challenge one is confronted with is that different communities have different environmental factors affecting the individual outcomes, and all individuals in a community share the same environment and intervention. In practice, data are often available from only a small number of communities, making it difficult if not impossible to adjust for these environmental confounders. In this paper we consider an extreme version of this dilemma, in which two communities each receives a different level of the intervention, and covariates and outcomes are measured on a random sample of independent individuals from each of the two populations; the results presented can be straightforwardly generalized to settings in which more than two communities are sampled. We address the question of what conditions are needed to estimate the causal effect of the intervention, defined in terms of an ideal experiment in which the exposed level of the intervention is assigned to both communities and individual outcomes are measured in the combined population, and then the clock is turned back and a control level of the intervention is assigned to both communities and individual outcomes are measured in the combined population. We refer to the difference in the expectation of these outcomes as the marginal (overall) treatment effect. We also discuss conditions needed for estimation of the treatment effect on the treated community. We apply a nonparametric structural equation model to define these causal effects and to establish conditions under which they are identified. These identifiability conditions provide guidance for the design of studies to investigate community level causal effects and for assessing the validity of causal interpretations when data are only available from a few communities. When the identifiability conditions fail to hold, the proposed statistical parameters still provide nonparametric treatment effect measures (albeit non-causal) whose statistical interpretations do not depend on model specifications. In addition, we study the use of a matched cohort sampling design in which the units of different communities are matched on individual factors. Finally, we provide semiparametric efficient and doubly robust targeted MLE estimators of the community level causal effect based on i.i.d. sampling and matched cohort sampling.
doi:10.1515/jci-2012-0011
PMCID: PMC4254657
PMID: 25485209
causal effect; causal effect among the treated; community-based intervention; efficient influence curve; environmental confounding
We would like to congratulate Lee, Nadler and Wasserman on their contribution to clustering and data reduction methods for high p and low n situations. A composite of clustering and traditional principal components analysis, treelets is an innovative method for multi-resolution analysis of unordered data. It is an improvement over traditional PCA and an important contribution to clustering methodology. Their paper presents theory and supporting applications addressing the two main goals of the treelet method: (1) Uncover the underlying structure of the data and (2) Data reduction prior to statistical learning methods. We will organize our discussion into two main parts to address their methodology in terms of each of these two goals. We will present and discuss treelets in terms of a clustering algorithm and an improvement over traditional PCA. We will also discuss the applicability of treelets to more general data, in particular, the application of treelets to microarray data.
doi:10.1214/07-AOAS137
PMCID: PMC4251495
PMID: 25478036
Summary
This article proposes resampling-based empirical Bayes multiple testing procedures for controlling a broad class of Type I error rates, defined as generalized tail probability (gTP) error rates, gTP(q, g) = Pr(g(Vn, Sn) > q), and generalized expected value (gEV) error rates, gEV(g) = E[g(Vn, Sn)], for arbitrary functions g(Vn, Sn) of the numbers of false positives Vn and true positives Sn. Of particular interest are error rates based on the proportion g(Vn, Sn) = Vn/(Vn + Sn) of Type I errors among the rejected hypotheses, such as the false discovery rate (FDR), FDR = E[Vn/(Vn + Sn)]. The proposed procedures offer several advantages over existing methods. They provide Type I error control for general data generating distributions, with arbitrary dependence structures among variables. Gains in power are achieved by deriving rejection regions based on guessed sets of true null hypotheses and null test statistics randomly sampled from joint distributions that account for the dependence structure of the data. The Type I error and power properties of an FDR-controlling version of the resampling-based empirical Bayes approach are investigated and compared to those of widely-used FDR-controlling linear step-up procedures in a simulation study. The Type I error and power trade-off achieved by the empirical Bayes procedures under a variety of testing scenarios allows this approach to be competitive with or outperform the Storey and Tibshirani (2003) linear step-up procedure, as an alternative to the classical Benjamini and Hochberg (1995) procedure.
doi:10.1002/bimj.200710473
PMCID: PMC4130579
PMID: 18932138
Adaptive; Adjusted p-value; Alternative hypothesis; Bootstrap; Correlation; Cut-off; Empirical Bayes; False discovery rate; Generalized expected value error rate; Generalized tail probability error rate; Joint distribution; Linear step-up procedure; Marginal procedure; Mixture model; Multiple hypothesis testing; Non-parametric; Null distribution; Null hypothesis; Posterior probability; Power; Prior probability; Proportion of true null hypotheses; q-value; R package; Receiver operator characteristic curve; Rejection region; Resampling; Simulation study; Software; t-statistic; Test statistic; Type I error rate
SUMMARY
In many randomized and observational studies the allocation of treatment among a sample of n independent and identically distributed units is a function of the covariates of all sampled units. As a result, the treatment labels among the units are possibly dependent, complicating estimation and posing challenges for statistical inference. For example, cluster randomized trials frequently sample communities from some target population, construct matched pairs of communities from those included in the sample based on some metric of similarity in baseline community characteristics, and then randomly allocate a treatment and a control intervention within each matched pair. In this case, the observed data can neither be represented as the realization of n independent random variables, nor, contrary to current practice, as the realization of n/2 independent random variables (treating the matched pair as the independent sampling unit). In this paper we study estimation of the average causal effect of a treatment under experimental designs in which treatment allocation potentially depends on the pre-intervention covariates of all units included in the sample. We define efficient targeted minimum loss based estimators for this general design, present a theorem that establishes the desired asymptotic normality of these estimators and allows for asymptotically valid statistical inference, and discuss implementation of these estimators. We further investigate the relative asymptotic efficiency of this design compared with a design in which unit-specific treatment assignment depends only on the units’ covariates. Our findings have practical implications for the optimal design and analysis of pair matched cluster randomized trials, as well as for observational studies in which treatment decisions may depend on characteristics of the entire sample.
PMCID: PMC4119765
PMID: 25097298
Cluster randomized trials; matching; asymptotic linearity of an estimator; causal effect; efficient influence curve; empirical process; confounding; dependent treatment allocation; G-computation formula; influence curve; loss function; adaptive randomization; semiparametric statistical model; targeted maximum likelihood estimation; targeted minimum loss based estimation (TMLE)
Objectives
To compare the performance of a targeted maximum likelihood estimator (TMLE) and a collaborative TMLE (CTMLE) to other estimators in a drug safety analysis, including a regression-based estimator, propensity score (PS)–based estimators, and an alternate doubly robust (DR) estimator in a real example and simulations.
Study Design and Setting
The real data set is a subset of observational data from Kaiser Permanente Northern California formatted for use in active drug safety surveillance. Both the real and simulated data sets include potential confounders, a treatment variable indicating use of one of two antidiabetic treatments and an outcome variable indicating occurrence of an acute myocardial infarction (AMI).
Results
In the real data example, there is no difference in AMI rates between treatments. In simulations, the double robustness property is demonstrated: DR estimators are consistent if either the initial outcome regression or PS estimator is consistent, whereas other estimators are inconsistent if the initial estimator is not consistent. In simulations with near-positivity violations, CTMLE performs well relative to other estimators by adaptively estimating the PS.
Conclusion
Each of the DR estimators was consistent, and TMLE and CTMLE had the smallest mean squared error in simulations.
doi:10.1016/j.jclinepi.2013.02.017
PMCID: PMC3818128
PMID: 23849159
Safety analysis; Targeted maximum likelihood estimation; Doubly robust; Causal inference; Collaborative targeted maximum likelihood estimation; Super learning
In randomized trials, investigators typically rely upon an unadjusted estimate of the mean outcome within each treatment arm to draw causal inferences. Statisticians have underscored the gain in efficiency that can be achieved from covariate adjustment in randomized trials with a focus on problems involving linear models. Despite recent theoretical advances, there has been a reluctance to adjust for covariates based on two primary reasons: (i) covariate-adjusted estimates based on conditional logistic regression models have been shown to be less precise and (ii) concern over the opportunity to manipulate the model selection process for covariate adjustments to obtain favorable results. In this paper, we address these two issues and summarize recent theoretical results on which is based a proposed general methodology for covariate adjustment under the framework of targeted maximum likelihood estimation in trials with two arms where the probability of treatment is 50%. The proposed methodology provides an estimate of the true causal parameter of interest representing the population-level treatment effect. It is compared with the estimates based on conditional logistic modeling, which only provide estimates of subgroup-level treatment effects rather than marginal (unconditional) treatment effects. We provide a clear criterion for determining whether a gain in efficiency can be achieved with covariate adjustment over the unadjusted method. We illustrate our strategy using a resampled clinical trial dataset from a placebo controlled phase 4 study. Results demonstrate that gains in efficiency can be achieved even with binary outcomes through covariate adjustment leading to increased statistical power.
doi:10.1002/sim.4301
PMCID: PMC4113477
PMID: 21751231
clinical trials; efficiency; covariate adjustment; variable selection
One of the identifiability assumptions of causal effects defined by marginal structural model (MSM) parameters is the experimental treatment assignment (ETA) assumption. Practical violations of this assumption frequently occur in data analysis when certain exposures are rarely observed within some strata of the population. The inverse probability of treatment weighted (IPTW) estimator is particularly sensitive to violations of this assumption; however, we demonstrate that this is a problem for all estimators of causal effects. This is due to the fact that the ETA assumption is about information (or lack thereof) in the data. A new class of causal models, causal models for realistic individualized exposure rules (CMRIER), is based on dynamic interventions. CMRIER generalize MSM, and their parameters remain fully identifiable from the observed data, even when the ETA assumption is violated, if the dynamic interventions are set to be realistic. Examples of such realistic interventions are provided. We argue that causal effects defined by CMRIER may be more appropriate in many situations, particularly those with policy considerations. Through simulation studies, we examine the performance of the IPTW estimator of the CMRIER parameters in contrast to that of the MSM parameters. We also apply the methodology to a real data analysis in air pollution epidemiology to illustrate the interpretation of the causal effects defined by CMRIER.
doi:10.1002/sim.4469
PMCID: PMC4113478
PMID: 22362629
causal inference; dynamic treatment regimes; IPTW estimator
The assumption of positivity or experimental treatment assignment requires that observed treatment levels vary within confounder strata. This article discusses the positivity assumption in the context of assessing model and parameter-specific identifiability of causal effects. Positivity violations occur when certain subgroups in a sample rarely or never receive some treatments of interest. The resulting sparsity in the data may increase bias with or without an increase in variance and can threaten valid inference. The parametric bootstrap is presented as a tool to assess the severity of such threats and its utility as a diagnostic is explored using simulated and real data. Several approaches for improving the identifiability of parameters in the presence of positivity violations are reviewed. Potential responses to data sparsity include restriction of the covariate adjustment set, use of an alternative projection function to define the target parameter within a marginal structural working model, restriction of the sample, and modification of the target intervention. All of these approaches can be understood as trading off proximity to the initial target of inference for identifiability; we advocate approaching this tradeoff systematically.
doi:10.1177/0962280210386207
PMCID: PMC4107929
PMID: 21030422
experimental treatment assignment; positivity; marginal structural model; inverse probability weight; double robust; causal inference; counterfactual; parametric bootstrap; realistic treatment rule; trimming; stabilised weights; truncation
SUMMARY
Researchers in clinical science and bioinformatics frequently aim to learn which of a set of candidate biomarkers is important in determining a given outcome, and to rank the contributions of the candidates accordingly. This article introduces a new approach to research questions of this type, based on targeted maximum-likelihood estimation of variable importance measures.
The methodology is illustrated using an example drawn from the treatment of HIV infection. Specifically, given a list of candidate mutations in the protease enzyme of HIV, we aim to discover mutations that reduce clinical virologic response to antiretroviral regimens containing the protease inhibitor lopinavir. In the context of this data example, the article reviews the motivation for covariate adjustment in the biomarker discovery process. A standard maximum-likelihood approach to this adjustment is compared with the targeted approach introduced here. Implementation of targeted maximum-likelihood estimation in the context of biomarker discovery is discussed, and the advantages of this approach are highlighted. Results of applying targeted maximum-likelihood estimation to identify lopinavir resistance mutations are presented and compared with results based on unadjusted mutation–outcome associations as well as results of a standard maximum-likelihood approach to adjustment.
The subset of mutations identified by targeted maximum likelihood as significant contributors to lopinavir resistance is found to be in better agreement with the current understanding of HIV antiretroviral resistance than the corresponding subsets identified by the other two approaches. This finding suggests that targeted estimation of variable importance represents a promising approach to biomarker discovery.
doi:10.1002/sim.3414
PMCID: PMC4107931
PMID: 18825650
biomarker discovery; variable importance; targeted maximum-likelihood estimation; HIV drug resistance
In a previously published article in this journal, Vansteeland et al. [Stat Methods Med Res. Epub ahead of print 12 November 2010. DOI: 10.1177/0962280210387717] address confounder selection in the context of causal effect estimation in observational studies. They discuss several selection strategies and propose a procedure whose performance is guided by the quality of the exposure effect estimator. The authors note that when a particular linearity condition is met, consistent estimation of the target parameter can be achieved even under dual misspecification of models for the association of confounders with exposure and outcome and demonstrate the performance of their procedure relative to other estimators when this condition holds. Our earlier published work on collaborative targeted minimum loss based learning provides a general theoretical framework for effective confounder selection that explains the findings of Vansteelandt et al. and underscores the appropriateness of their suggestions that a confounder selection procedure should be concerned with directly targeting the quality of the estimate and that desirable estimators produce valid confidence intervals and are robust to dual misspecification.
doi:10.1177/0962280212437451
PMCID: PMC4081493
PMID: 22368176
collaborative double robustness; TMLE; collaborative targeted maximum likelihood estimation; propensity score; confounder selection; causal effect; causal inference; dual misspecification
Summary
The natural direct effect (NDE), or the effect of an exposure on an outcome if an
intermediate variable was set to the level it would have been in the absence of the exposure, is
often of interest to investigators. In general, the statistical parameter associated with the NDE is
difficult to estimate in the non-parametric model, particularly when the intermediate variable is
continuous or high dimensional. In this paper we introduce a new causal parameter called the natural
direct effect among the untreated, discus identifiability assumptions, propose a sensitivity
analysis for some of the assumptions, and show that this new parameter is equivalent to the NDE in a
randomized controlled trial. We also present a targeted minimum loss estimator (TMLE), a locally
efficient, double robust substitution estimator for the statistical parameter associated with this
causal parameter. The TMLE can be applied to problems with continuous and high dimensional
intermediate variables, and can be used to estimate the NDE in a randomized controlled trial with
such data. Additionally, we define and discuss the estimation of three related causal parameters:
the natural direct effect among the treated, the indirect effect among the untreated and the
indirect effect among the treated.
doi:10.1111/biom.12022
PMCID: PMC3692606
PMID: 23607645
Causal inference; direct effect; indirect effect; mediation analysis; semiparametric models; targeted minimum loss estimation
Summary
Despite modern effective HIV treatment, hepatitis C virus (HCV) co-infection is associated with a high risk of progression to end-stage liver disease (ESLD) which has emerged as the primary cause of death in this population. Clinical interest lies in determining the impact of clearance of HCV on risk for ESLD. In this case study, we examine whether HCV clearance affects risk of ESLD using data from the multicenter Canadian Co-infection Cohort Study. Complications in this survival analysis arise from the time-dependent nature of the data, the presence of baseline confounders, loss to follow-up, and confounders that change over time, all of which can obscure the causal effect of interest. Additional challenges included non-censoring variable missingness and event sparsity.
In order to efficiently estimate the ESLD-free survival probabilities under a specific history of HCV clearance, we demonstrate the doubly-robust and semiparametric efficient method of Targeted Maximum Likelihood Estimation (TMLE). Marginal structural models (MSM) can be used to model the effect of viral clearance (expressed as a hazard ratio) on ESLD-free survival and we demonstrate a way to estimate the parameters of a logistic model for the hazard function with TMLE. We show the theoretical derivation of the efficient influence curves for the parameters of two different MSMs and how they can be used to produce variance approximations for parameter estimates. Finally, the data analysis evaluating the impact of HCV on ESLD was undertaken using multiple imputations to account for the non-monotone missing data.
doi:10.1111/biom.12105
PMCID: PMC3954273
PMID: 24571372
Double-robust; Inverse probability of treatment weighting; Kaplan-Meier; Longitudinal data; Marginal structural model; Survival analysis; Targeted maximum likelihood estimation
Abstract
The Tshepo study was the first clinical trial to evaluate outcomes of adults receiving nevirapine (NVP)-based versus efavirenz (EFV)-based combination antiretroviral therapy (cART) in Botswana. This was a 3 year study (n=650) comparing the efficacy and tolerability of various first-line cART regimens, stratified by baseline CD4+: <200 (low) vs. 201-350 (high). Using targeted maximum likelihood estimation (TMLE), we retrospectively evaluated the causal effect of assigned NNRTI on time to virologic failure or death [intent-to-treat (ITT)] and time to minimum of virologic failure, death, or treatment modifying toxicity [time to loss of virological response (TLOVR)] by sex and baseline CD4+. Sex did significantly modify the effect of EFV versus NVP for both the ITT and TLOVR outcomes with risk differences in the probability of survival of males versus the females of approximately 6% (p=0.015) and 12% (p=0.001), respectively. Baseline CD4+ also modified the effect of EFV versus NVP for the TLOVR outcome, with a mean difference in survival probability of approximately 12% (p=0.023) in the high versus low CD4+ cell count group. TMLE appears to be an efficient technique that allows for the clinically meaningful delineation and interpretation of the causal effect of NNRTI treatment and effect modification by sex and baseline CD4+ cell count strata in this study. EFV-treated women and NVP-treated men had more favorable cART outcomes. In addition, adults initiating EFV-based cART at higher baseline CD4+ cell count values had more favorable outcomes compared to those initiating NVP-based cART.
doi:10.1089/aid.2011.0349
PMCID: PMC3423643
PMID: 22309114
We define a new measure of variable importance of an exposure on a continuous outcome, accounting for potential confounders. The exposure features a reference level x0 with positive mass and a continuum of other levels. For the purpose of estimating it, we fully develop the semi-parametric estimation methodology called targeted minimum loss estimation methodology (TMLE) [23, 22]. We cover the whole spectrum of its theoretical study (convergence of the iterative procedure which is at the core of the TMLE methodology; consistency and asymptotic normality of the estimator), practical implementation, simulation study and application to a genomic example that originally motivated this article. In the latter, the exposure X and response Y are, respectively, the DNA copy number and expression level of a given gene in a cancer cell. Here, the reference level is x0 = 2, that is the expected DNA copy number in a normal cell. The confounder is a measure of the methylation of the gene. The fact that there is no clear biological indication that X and Y can be interpreted as an exposure and a response, respectively, is not problematic.
doi:10.1214/12-EJS703
PMCID: PMC3546832
PMID: 23336014
Variable importance measure; non-parametric estimation; targeted minimum loss estimation; robustness; asymptotics
A new class of Marginal Structural Models (MSMs), History-Restricted MSMs (HRMSMs), was recently introduced for longitudinal data for the purpose of defining causal parameters which may often be better suited for public health research or at least more practicable than MSMs (6, 2). HRMSMs allow investigators to analyze the causal effect of a treatment on an outcome based on a fixed, shorter and user-specified history of exposure compared to MSMs. By default, the latter represent the treatment causal effect of interest based on a treatment history defined by the treatments assigned between the study’s start and outcome collection. We lay out in this article the formal statistical framework behind HRMSMs. Beyond allowing a more flexible causal analysis, HRMSMs improve computational tractability and mitigate statistical power concerns when designing longitudinal studies. We also develop three consistent estimators of HRMSM parameters under sufficient model assumptions: the Inverse Probability of Treatment Weighted (IPTW), G-computation and Double Robust (DR) estimators. In addition, we show that the assumptions commonly adopted for identification and consistent estimation of MSM parameters (existence of counterfactuals, consistency, time-ordering and sequential randomization assumptions) also lead to identification and consistent estimation of HRMSM parameters.
doi:10.1214/07-EJS050
PMCID: PMC3475192
PMID: 23087778
causal inference; counterfactual; marginal structural model; longitudinal study; IPTW; G-computation; Double Robust
Background:
The evidence for the effectiveness of antihypertensive medication use for slowing decline in kidney function in older persons is sparse. We addressed this research question by the application of novel methods in a marginal structural model.
Methods:
Change in kidney function was measured by two or more measures of cystatin C in 1,576 hypertensive participants in the Cardiovascular Health Study over 7 years of follow-up (1989–1997 in four U.S. communities). The exposure of interest was antihypertensive medication use. We used a novel estimator in a marginal structural model to account for bias due to confounding and informative censoring.
Results:
The mean annual decline in eGFR was 2.41 ± 4.91 mL/min/1.73 m2. In unadjusted analysis, antihypertensive medication use was not associated with annual change in kidney function. Traditional multivariable regression did not substantially change these estimates. Based on a marginal structural analysis, persons on antihypertensives had slower declines in kidney function; participants had an estimated 0.88 (0.13, 1.63) ml/min/1.73 m2 per year slower decline in eGFR compared with persons on no treatment. In a model that also accounted for bias due to informative censoring, the estimate for the treatment effect was 2.23 (−0.13, 4.59) ml/min/1.73 m2 per year slower decline in eGFR.
Conclusion:
In summary, estimates from a marginal structural model suggested that antihypertensive therapy was associated with preserved kidney function in hypertensive elderly adults. Confirmatory studies may provide power to determine the strength and validity of the findings.
doi:10.2202/1557-4679.1320
PMCID: PMC3204667
PMID: 22049266
aged; kidney function; hypertension; marginal structural model
There is an active debate in the literature on censored data about the relative performance of model based maximum likelihood estimators, IPCW-estimators, and a variety of double robust semiparametric efficient estimators. Kang and Schafer (2007) demonstrate the fragility of double robust and IPCW-estimators in a simulation study with positivity violations. They focus on a simple missing data problem with covariates where one desires to estimate the mean of an outcome that is subject to missingness. Responses by Robins, et al. (2007), Tsiatis and Davidian (2007), Tan (2007) and Ridgeway and McCaffrey (2007) further explore the challenges faced by double robust estimators and offer suggestions for improving their stability. In this article, we join the debate by presenting targeted maximum likelihood estimators (TMLEs). We demonstrate that TMLEs that guarantee that the parametric submodel employed by the TMLE procedure respects the global bounds on the continuous outcomes, are especially suitable for dealing with positivity violations because in addition to being double robust and semiparametric efficient, they are substitution estimators. We demonstrate the practical performance of TMLEs relative to other estimators in the simulations designed by Kang and Schafer (2007) and in modified simulations with even greater estimation challenges.
doi:10.2202/1557-4679.1308
PMCID: PMC3173607
PMID: 21931570
censored data; collaborative double robustness; collaborative targeted maximum likelihood estimation; double robust; estimator selection; inverse probability of censoring weighting; locally efficient estimation; maximum likelihood estimation; semiparametric model; targeted maximum likelihood estimation; targeted minimum loss based estimation; targeted nuisance parameter estimator selection
Quantitative trait loci mapping is focused on identifying the positions and effect of genes underlying an an observed trait. We present a collaborative targeted maximum likelihood estimator in a semi-parametric model using a newly proposed 2-part super learning algorithm to find quantitative trait loci genes in listeria data. Results are compared to the parametric composite interval mapping approach.
doi:10.1016/j.spl.2010.11.001
PMCID: PMC3090625
PMID: 21572586
collaborative targeted maximum likelihood estimation; quantitative trait loci; super learner; machine learning
The Cox proportional hazards model or its discrete time analogue, the logistic failure time model, posit highly restrictive parametric models and attempt to estimate parameters which are specific to the model proposed. These methods are typically implemented when assessing effect modification in survival analyses despite their flaws. The targeted maximum likelihood estimation (TMLE) methodology is more robust than the methods typically implemented and allows practitioners to estimate parameters that directly answer the question of interest. TMLE will be used in this paper to estimate two newly proposed parameters of interest that quantify effect modification in the time to event setting. These methods are then applied to the Tshepo study to assess if either gender or baseline CD4 level modify the effect of two cART therapies of interest, efavirenz (EFV) and nevirapine (NVP), on the progression of HIV. The results show that women tend to have more favorable outcomes using EFV while males tend to have more favorable outcomes with NVP. Furthermore, EFV tends to be favorable compared to NVP for individuals at high CD4 levels.
doi:10.2202/1557-4679.1307
PMCID: PMC3083138
PMID: 21556287
causal effect; semi-parametric; censored longitudinal data; double robust; efficient influence curve; influence curve; G-computation; Targeted Maximum Likelihood Estimation; Cox-proportional hazards; survival analysis
We consider two-stage sampling designs, including so-called nested case control studies, where one takes a random sample from a target population and completes measurements on each subject in the first stage. The second stage involves drawing a subsample from the original sample, collecting additional data on the subsample. This data structure can be viewed as a missing data structure on the full-data structure collected in the second-stage of the study. Methods for analyzing two-stage designs include parametric maximum likelihood estimation and estimating equation methodology. We propose an inverse probability of censoring weighted targeted maximum likelihood estimator (IPCW-TMLE) in two-stage sampling designs and present simulation studies featuring this estimator.
doi:10.2202/1557-4679.1217
PMCID: PMC3083136
PMID: 21556285
two-stage designs; targeted maximum likelihood estimators; nested case control studies; double robust estimation
In longitudinal and repeated measures data analysis, often the goal is to determine the effect of a treatment or aspect on a particular outcome (e.g., disease progression). We consider a semiparametric repeated measures regression model, where the parametric component models effect of the variable of interest and any modification by other covariates. The expectation of this parametric component over the other covariates is a measure of variable importance. Here, we present a targeted maximum likelihood estimator of the finite dimensional regression parameter, which is easily estimated using standard software for generalized estimating equations.
The targeted maximum likelihood method provides double robust and locally efficient estimates of the variable importance parameters and inference based on the influence curve. We demonstrate these properties through simulation under correct and incorrect model specification, and apply our method in practice to estimating the activity of transcription factor (TF) over cell cycle in yeast. We specifically target the importance of SWI4, SWI6, MBP1, MCM1, ACE2, FKH2, NDD1, and SWI5.
The semiparametric model allows us to determine the importance of a TF at specific time points by specifying time indicators as potential effect modifiers of the TF. Our results are promising, showing significant importance trends during the expected time periods. This methodology can also be used as a variable importance analysis tool to assess the effect of a large number of variables such as gene expressions or single nucleotide polymorphisms.
doi:10.2202/1544-6115.1553
PMCID: PMC3122882
PMID: 21291412
targeted maximum likelihood; semiparametric; repeated measures; longitudinal; transcription factors
Targeted maximum likelihood estimation of a parameter of a data generating distribution, known to be an element of a semi-parametric model, involves constructing a parametric model through an initial density estimator with parameter ɛ representing an amount of fluctuation of the initial density estimator, where the score of this fluctuation model at ɛ = 0 equals the efficient influence curve/canonical gradient. The latter constraint can be satisfied by many parametric fluctuation models since it represents only a local constraint of its behavior at zero fluctuation. However, it is very important that the fluctuations stay within the semi-parametric model for the observed data distribution, even if the parameter can be defined on fluctuations that fall outside the assumed observed data model. In particular, in the context of sparse data, by which we mean situations where the Fisher information is low, a violation of this property can heavily affect the performance of the estimator. This paper presents a fluctuation approach that guarantees the fluctuated density estimator remains inside the bounds of the data model. We demonstrate this in the context of estimation of a causal effect of a binary treatment on a continuous outcome that is bounded. It results in a targeted maximum likelihood estimator that inherently respects known bounds, and consequently is more robust in sparse data situations than the targeted MLE using a naive fluctuation model.
When an estimation procedure incorporates weights, observations having large weights relative to the rest heavily influence the point estimate and inflate the variance. Truncating these weights is a common approach to reducing the variance, but it can also introduce bias into the estimate. We present an alternative targeted maximum likelihood estimation (TMLE) approach that dampens the effect of these heavily weighted observations. As a substitution estimator, TMLE respects the global constraints of the observed data model. For example, when outcomes are binary, a fluctuation of an initial density estimate on the logit scale constrains predicted probabilities to be between 0 and 1. This inherent enforcement of bounds has been extended to continuous outcomes. Simulation study results indicate that this approach is on a par with, and many times superior to, fluctuating on the linear scale, and in particular is more robust when there is sparsity in the data.
doi:10.2202/1557-4679.1260
PMCID: PMC3126669
PMID: 21731529
targeted maximum likelihood estimation; TMLE; causal effect