Cancer Epidemiology Unit, Department of Medical Sciences, University of Turin and CPO Piedmont, Via Santena 7, Turin, 10126, Italy.
Department of Medical Sciences, University of Turin, Turin, Italy.
Eur J Epidemiol. 2024 Oct;39(10):1097-1108. doi: 10.1007/s10654-024-01173-x. Epub 2024 Nov 13.
In causal inference, parametric models are usually employed to address causal questions estimating the effect of interest. However, parametric models rely on the correct model specification assumption that, if not met, leads to biased effect estimates. Correct model specification is challenging, especially in high-dimensional settings. Incorporating Machine Learning (ML) into causal analyses may reduce the bias arising from model misspecification, since ML methods do not require the specification of a functional form of the relationship between variables. However, when ML predictions are directly plugged in a predefined formula of the effect of interest, there is the risk of introducing a "plug-in bias" in the effect measure. To overcome this problem and to achieve useful asymptotic properties, new estimators that combine the predictive potential of ML and the ability of traditional statistical methods to make inference about population parameters have been proposed. For epidemiologists interested in taking advantage of ML for causal inference investigations, we provide an overview of three estimators that represent the current state-of-art, namely Targeted Maximum Likelihood Estimation (TMLE), Augmented Inverse Probability Weighting (AIPW) and Double/Debiased Machine Learning (DML).
在因果推断中，通常采用参数模型来解决因果问题，估计感兴趣的效应。然而，参数模型依赖于正确的模型规格假设，如果不满足，会导致有偏的效应估计。正确的模型规格是具有挑战性的，特别是在高维环境中。将机器学习（ML）纳入因果分析中，可以减少因模型不规范而产生的偏差，因为 ML 方法不需要指定变量之间关系的函数形式。然而，当 ML 预测值直接插入到感兴趣的效应的预定义公式中时，可能会在效应量测中引入“插件偏差”。为了克服这个问题并实现有用的渐近性质，已经提出了新的估计量，它们结合了 ML 的预测潜力和传统统计方法对总体参数进行推断的能力。对于有兴趣利用 ML 进行因果推断研究的流行病学家，我们提供了三种当前最先进的估计量的概述，即靶向最大似然估计（TMLE）、增强逆概率加权（AIPW）和双重/去偏机器学习（DML）。
