As discussed above, the double robustness property of TMLE indicates that the estimator will be consistent if either component is consistently estimated, and efficient if both are consistently estimated.
Previous theoretical and empirical papers, including those with simulation studies, have demonstrated that TMLE will often outperform parametric regression when the regression is misspecified (e.g., omitted variable bias) and in high-dimensional settings with collinear variables (Bembom et al.
The TMLE for each A was calculated separately, and influence curve-based estimates were used to calculate standard errors and corresponding p-values, adjusting for multiple testing using the false discovery rate at an 0.05 level (Benjamini and Hochberg 1995) as in previous work (Wang et al.
"tmle: An R Package for Targeted Maximum Likelihood Estimation." Journal of Statistical Software 51: 1-35.
(37), the most common genes observed in the Turkish population in non-syndromic autosomal recessive hearing loss following the GJB2 gene were reported and the OTOF gene was reported to be the fourth relevant gene with mutations with a frequency of 5%, following MYO15A (9.9%), TMlE
(6.6%), and TMC1 (6.6%).
The marginal geometric mean difference of p,p'-DDT/DDE serum concentrations for each intervention was evaluated in separate models using targeted maximum likelihood estimations (TMLE), a doubly robust substitution estimator that generates unbiased estimates if either models for the estimation of the exposure [E(Y|A,W)] or determinant mechanisms [E(A|W)] are correct (Rose and van der Laan 2011; van der Laan 2006; van der Laan and Rubin 2006).
In addition, the use of TMLE using data-adaptive machine learning algorithms allowed for a targeted approach, where the research question defines the statistical analysis, to understand the effect of interventions on DDT/DDE body burden, without a priori specifying a model's input or making assumptions about the form of the respective relationships among exposure, outcome, and covariates.
TMLE is a double robust semiparametric efficient estimation method, and is tailored to minimize bias and maximize precision as proven by theory (Chambaz et al.
TMLE is an optimal way to perform detailed mediation analysis.