Double robust semi-supervised inference for the mean: selection bias under MAR labeling with decaying overlap
Double robust semi-supervised inference for the mean: selection bias under MAR labeling with decaying overlap
复制标题
均值的双重鲁棒半监督推理：具有衰减重叠的 MAR 标签下的选择偏差
DOI：
10.1093/imaiai/iaad021
复制
发表时间：
2023
期刊：
影响因子：
--
通讯作者：
Bradic, Jelena
中科院分区：
文献类型：
--
作者：
Zhang, Yuqian;Chakrabortty, Abhishek;Bradic, Jelena
半监督推理（Semi-supervised inference，SS）近年来受到了广泛的关注。除了中等大小的标记数据，SS设置的特征在于附加的、更大的未标记数据。的设置，使SS推理独特和不同的标准缺失数据问题，由于自然违反所谓的“积极性”或“重叠”的假设。然而，大多数SS文献都隐含地要求平均分布，即，标签中没有选择偏差。随机缺失类型标签的推理挑战允许选择偏倚，不可避免地会因倾向评分（PS）的衰减性质而加剧。我们解决这个差距的一个原型问题，估计响应的平均值。本文提出了一种双稳健SS均值估计，并给出了其渐近性质的完整刻画。只要结果或PS模型被正确指定，所提出的估计量是一致的。当两个模型都被正确指定时，我们提供的推理结果具有非标准的一致性率，该一致性率取决于较小的大小。结果也扩展到因果推理与不平衡的治疗组。此外，我们提供了几个新的选择的模型和估计的衰减PS，包括一个新的偏移逻辑模型和分层标签模型。我们提出了他们的高，低维设置下的属性。这些可能是独立的利益。最后，我们提出了广泛的模拟，也是一个真实的数据应用。
Semi-supervised (SS) inference has received much attention in recent years. Apart from a moderate-sized labeled data,, the SS setting is characterized by an additional,much larger sized, unlabeled data,. The setting of, makes SS inference unique and different from the standard missing data problems, owing to natural violation of the so-called ‘positivity’ or ‘overlap’ assumption. However, most of the SS literature implicitly assumesandto be equally distributed, i.e., no selection bias in the labeling. Inferential challenges in missing at random type labeling allowing for selection bias, are inevitably exacerbated by the decaying nature of the propensity score (PS). We address this gap for a prototype problem, the estimation of the response’s mean. We propose a double robust SS mean estimator and give a complete characterization of its asymptotic properties. The proposed estimator is consistent as long as either the outcome or the PS model is correctly specified. When both models are correctly specified, we provide inference results with a non-standard consistency rate that depends on the smaller size. The results are also extended to causal inference with imbalanced treatment groups. Further, we provide several novel choices of models and estimators of the decaying PS, including a novel offset logistic model and a stratified labeling model. We present their properties under both high- and low-dimensional settings. These may be of independent interest. Lastly, we present extensive simulations and also a real data application.
登录
查看更多内容
影响因子：
4.5
作者：
Chakrabortty, Abhishek;Cai, Tianxi
通讯作者：
Cai, Tianxi
影响因子：
4.5
作者：
Zhang, Anru;Brown, Lawrence D.;Cai, T. Tony
通讯作者：
Cai, T. Tony
影响因子：
1.9
作者：
Ertefaie, Ashkan;Hejazi, Nima S.;van der Laan, Mark J.
通讯作者：
van der Laan, Mark J.
协变量平移问题中的半监督线性回归
DOI：
10.5555/2789272.2912101
发表时间：
2015
期刊：
J. Mach. Learn. Res.
影响因子：
--
作者：
K. J. Ryan;M. Culp
通讯作者：
M. Culp
DOI：
10.1111/rssb.12357
发表时间：
2020-01-20
影响因子：
5.8
作者：
Cai, T. Tony;Guo, Zijian
通讯作者：
Guo, Zijian