A Density Ratio Super Learner
This work addresses a key challenge in causal inference for statisticians, but it appears incremental as it builds on existing super learning methods with a new loss function.
The authors tackled the problem of estimating density ratios, which are important in causal inference, by developing an ensemble estimator with a novel loss function based on super learning. They demonstrated its performance through simulations in mediation analysis and longitudinal modified treatment policy, showing empirical results.
The estimation of the ratio of two density probability functions is of great interest in many statistics fields, including causal inference. In this study, we develop an ensemble estimator of density ratios with a novel loss function based on super learning. We show that this novel loss function is qualified for building super learners. Two simulations corresponding to mediation analysis and longitudinal modified treatment policy in causal inference, where density ratios are nuisance parameters, are conducted to show our density ratio super learner's performance empirically.