针对异质性总体的倾向性评分加权方法比较研究*

王旭晴 ,  胡禹 ,  秦国友

中国卫生统计 ›› 2026, Vol. 43 ›› Issue (5) : 714 -720.

PDF (1102KB)
中国卫生统计 ›› 2026, Vol. 43 ›› Issue (5) : 714 -720. DOI: 10.11783/j.issn.1002-3674.2026.05.013
应用研究

针对异质性总体的倾向性评分加权方法比较研究*

作者信息 +

Author information +
文章历史 +
PDF (1128K)

摘要

目的 探讨在观察性研究中,五种倾向性评分加权方法估计异质性总体平均处理效应的性能。方法 采用Monte Carlo 模拟方法,设置了六种异质性场景以及三种倾向性评分分布重叠程度,生成相应数据集,再将稳定的逆概率加权、修剪的逆概率加权、重叠加权、经验似然加权以及协变量平衡倾向性评分加权估计器分别应用于估计平均处理效应,并根据相对偏差和相对均方根误差两个指标来衡量各估计器的性能。结果 在异质性总体中,模型正确指定的情况下,推断目标为整个人群的估计器相对偏差明显小于其他估计器,其中协变量平衡倾向性评分加权估计器表现最佳,其相对均方根误差始终最大。结论 在估计异质性总体的平均处理效应时,重叠加权方法可能存在较大偏倚,协变量平衡倾向性评分加权方法在准确性和效率等方面均相对最优。

关键词

倾向性评分 / 平均处理效应 / 异质性总体 / 加权

Key words

引用本文

引用格式 ▾
王旭晴,胡禹,秦国友. 针对异质性总体的倾向性评分加权方法比较研究*[J]. 中国卫生统计, 2026, 43(5): 714-720 DOI:10.11783/j.issn.1002-3674.2026.05.013

登录浏览全文

4963

注册一个新账户 忘记密码

参考文献

[1]

Hernán MA,Robins JM. Causal Inference: What If[M]. Boca Raton: Chapman & Hall/CRC,2020.

[2]

孙婷,秦国友,武振宇, 等. 不同混杂结构条件下各倾向性评分方法的模拟比较研究[J].中国卫生统计,2017,34(3):415-420.

[3]

Crump RK,Hotz VJ,Imbens GW,et al. Dealing with limited overlap in estimation of average treatment effects[J].Biometrika,2009,96(1):187-199.

[4]

Imai K,Ratkovic M. Covariate balancing propensity score[J].Journal of the Royal Statistical Society Series B, Statistical methodology,2014,76(1):243-263.

[5]

Han P,Wang L. Estimation with missing data: beyond double robustness[J].Biometrika,2013,100(2):417-430.

[6]

Liu Y,Fan Y. Biased—sample empirical likelihood weighting for missing data problems: an alternative to inverse probability weighting[J].Journal of the Royal Statistical Society Series B: Statistical Methodology,2023,85(1):67-83.

[7]

Desai RJ,Franklin JM. Alternative approaches for confounding adjustment in observational studies using weighting based on the propensity score: a primer for practitioners[J].BMJ,2019,367:l5657—l5657.

[8]

涂博祥,秦婴逸,吴骋, 等. 倾向性评分加权方法介绍及R 软件实现[J].中国循证医学杂志,2022,22(3):365-372.

[9]

Zagar AJ,Kadziola Z,Lipkovich I,et al. Evaluating different strategies for estimating treatment effects in observational studies[J].Journal of Biopharmaceutical Statistics,2017,27(3):535-553.

[10]

Li F,Morgan KL,Zaslavsky AM. Balancing covariates via propensity score weighting[J].Journal of the American Statistical Association,2018,113(521):390-400.

[11]

Brooks JM,Chapman CG,Schroeder MC. Understanding treatment effect estimates when treatment effects are heterogeneous for more than one outcome[J].Applied Health Economics and Health Policy,2018,16(3):381-393.

[12]

Robertson SE,Leith A,Schmid CH,et al. Assessing heterogeneity of treatment effects in observational studies[J].American Journal of Epidemiology,2021,190(6):1088-1100.

[13]

Dahabreh IJ,Hayward R,Kent DM. Using group data to treat individuals: understanding heterogeneous treatment effects in the age of precision medicine and patient—centred evidence[J].International Journal of Epidemiology,2016,45(6):2184-2193.

[14]

Rubin DB. Estimating causal effects from large data sets using propensity scores[J].Annals of Internal Medicine,1997,127(8 Part 2):757-763.

[15]

Li F,Thomas LE. Addressing extreme propensity scores via the overlap weights[J].American Journal of Epidemiology,2019,188(1):250-257.

[16]

Hainmueller J. Entropy balancing for causal effects: A multivariate reweighting method to produce balanced samples in observational studies[J].Political Analysis, 2012, 20(1): 25-46.

[17]

Kereiakes DJ,Obenchain RL,Barber BL,et al. Abciximab provides cost—effective survival advantage in high—volume interventional practice[J].American Heart Journal,2000,140(4):603-610.

[18]

Han P. A further study of the multiply robust estimator in missing data analysis[J].Journal of Statistical Planning and Inference,2014,148:101-110.

[19]

Zhou YJ,Matsouaka RA,Thomas L. Propensity score weighting under limited overlap and model misspecification[J].Statistical Methods in Medical Research,2020,29(12):3721-3756.

[20]

Garrido MM,Lum J,Pizer SD. Vector—based kernel weighting: A simple estimator for improving precision and bias of average treatment effects in multiple treatment settings[J].Statistics in Medicine,2021,40(5):1204-1223.

[21]

Yang S, Lin S, Li N, et al. Burden, trends, and risk factors of esophageal cancer in China from 1990 to 2017: an up—to—date overview and comparison with those in Japan and South Korea[J].Journal of Hematology & Oncology, 2020, 13(1): 146.

[22]

Shigaki H, Baba Y, Watanabe M,et al. LINE—1 hypomethylation in noncancerous esophageal mucosae is associated with smoking history[J].Annals of Surgical Oncology, 2012, 19(13): 4238-4243.

[23]

Jin EH, Han K, Shin CM, et al. Sex and Tumor—Site Differences in the Association of Alcohol Intake With the Risk of Early—Onset Colorectal Cancer[J].Journal of Clinical Oncology: Official Journal of the American Society of Clinical Oncology, 2023, 41(22): 3816-3825.

[24]

国务院第七次全国人口普查领导小组办公室. 中国人口普查年鉴2020[M].北京:中国统计出版社, 2022.

[25]

地力亚尔·吾斯曼江. 1990—2019年归因于吸烟、饮酒、高BMI的中国食管癌死亡率趋势分析[D].乌鲁木齐: 新疆医科大学, 2023.

[26]

陈思念, 曾智, 朱嘉雯, 等. 中国归因于饮酒的食管癌疾病负担分析与预测:基于1990—2019 年中国疾病负担的年龄—时期—队列研究[J].肿瘤, 2024, 44(2): 158-168.

基金资助

上海市市级科技重大专项(ZD2021CY001)

国家自然科学基金(82473724)

AI Summary AI Mindmap
PDF (1102KB)

0

访问

0

被引

详细

导航
相关文章

AI思维导图

/

〈 〉