Missing data imputation and synthetic data simulation through modeling graphical probabilistic dependencies between variables (ModGraProDep): An application to breast cancer survival

作者：

Highlights：

• ModGraProDep addresses missing data and small sample size issues in survival analysis.

• It provides 4 models which could outperform useful algorithms for these issues.

• It also allows for identifying probability dependencies between variables.

• Saturated models should be discarded for any imputation/data simulation task.

摘要

•ModGraProDep addresses missing data and small sample size issues in survival analysis.•It provides 4 models which could outperform useful algorithms for these issues.•It also allows for identifying probability dependencies between variables.•Saturated models should be discarded for any imputation/data simulation task.

论文关键词：Breast cancer,Survival,Graphical models,Missing data,Oversampling,Simulation

论文评审过程：Received 31 October 2019, Revised 12 February 2020, Accepted 2 May 2020, Available online 24 May 2020, Version of Record 12 June 2020.

论文官网地址：https://doi.org/10.1016/j.artmed.2020.101875