Showing 1–2 of 2 results for author: Bong, S
-
Local causal effects with continuous exposures: A matching estimator for the average causal derivative effect
Authors:
Suhwan Bong,
Kwonsang Lee
Abstract:
The estimation of causal effects is a fundamental goal in the field of causal inference. However, it is challenging for various reasons. One reason is that the exposure (or treatment) is naturally continuous in many real-world scenarios. When dealing with continuous exposure, dichotomizing the exposure variable based on a pre-defined threshold may result in a biased understanding of causal relatio…
▽ More
The estimation of causal effects is a fundamental goal in the field of causal inference. However, it is challenging for various reasons. One reason is that the exposure (or treatment) is naturally continuous in many real-world scenarios. When dealing with continuous exposure, dichotomizing the exposure variable based on a pre-defined threshold may result in a biased understanding of causal relationships. In this paper, we propose a novel causal inference framework that can measure the causal effect of continuous exposure. We define the expectation of a derivative of potential outcomes at a specific exposure level as the average causal derivative effect. Additionally, we propose a matching method for this estimator and propose a permutation approach to test the hypothesis of no local causal effect. We also investigate the asymptotic properties of the proposed estimator and examine its performance through simulation studies. Finally, we apply this causal framework in a real data example of Chronic Obstructive Pulmonary Disease (COPD) patients.
△ Less
Submitted 30 November, 2023;
originally announced November 2023.
-
Differential recall bias in estimating treatment effects in observational studies
Authors:
Suhwan Bong,
Kwonsang Lee,
Francesca Dominici
Abstract:
Observational studies are frequently used to estimate the effect of an exposure or treatment on an outcome. To obtain an unbiased estimate of the treatment effect, it is crucial to measure the exposure accurately. A common type of exposure misclassification is recall bias, which occurs in retrospective cohort studies when study subjects may inaccurately recall their past exposure. Particularly cha…
▽ More
Observational studies are frequently used to estimate the effect of an exposure or treatment on an outcome. To obtain an unbiased estimate of the treatment effect, it is crucial to measure the exposure accurately. A common type of exposure misclassification is recall bias, which occurs in retrospective cohort studies when study subjects may inaccurately recall their past exposure. Particularly challenging is differential recall bias in the context of self-reported binary exposures, where the bias may be directional rather than random , and its extent varies according to the outcomes experienced. This paper makes several contributions: (1) it establishes bounds for the average treatment effect (ATE) even when a validation study is not available; (2) it proposes multiple estimation methods across various strategies predicated on different assumptions; and (3) it suggests a sensitivity analysis technique to assess the robustness of the causal conclusion, incorporating insights from prior research. The effectiveness of these methods is demonstrated through simulation studies that explore various model misspecification scenarios. These approaches are then applied to investigate the effect of childhood physical abuse on mental health in adulthood.
△ Less
Submitted 30 June, 2024; v1 submitted 5 July, 2023;
originally announced July 2023.
