2013/11/06 by Ashkan Ertefaie, Ertefaie, Ashkan, Masoud Asgharian +4
Mathematics · #Advanced Causal Inference Techniques #FOS: Mathematics #Statistical Methods and Bayesian Inference #Statistical Methods and Inference #Statistics Theory (math.ST)
paper · pdf · doi:10.48550/arxiv.1311.1400
openalex publication_date 2013/11/06 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
The pervasive use of prevalent cohort studies on disease duration,\nincreasingly calls for appropriate methodologies to account for the biases that\ninvariably accompany samples formed by such data. It is well-known, for\nexample, that subjects with shorter lifetime are less likely to be present in\nsuch studies. Moreover, certain covariate values could be preferentially\nselected into the sample, being linked to the long-term survivors. The existing\nmethodology for estimation of the propensity score using data collected on\nprevalent cases requires the correct conditional survival/hazard function given\nthe treatment and covariates. This requirement can be alleviated if the disease\nunder study has stationary incidence, the so-called stationarity assumption. We\npropose a nonparametric adjustment technique based on a weighted estimating\nequation for estimating the propensity score which does not require modeling\nthe conditional survival/hazard function when the stationarity assumption\nholds. Large sample properties of the estimator is established and its small\nsample behavior is studied via simulation.\n