Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


ThesisDanqiZhu.pdf (344.95 KB)

ETD Abstract Container

Abstract Header

Evaluating causal effect in time-to-event observarional data with propensity score matching

Zhu, Danqi
http://rave.ohiolink.edu/etdc/view?acc_num=osu1449767366

Abstract Details

2016, Master of Science, Ohio State University, Public Health.
In observational time-to-event data, traditional survival data analysis techniques such as the log-rank test and the Cox proportional hazards model may introduce biased results. Matching based on propensity scores are promising to conduct causal inference. Given the matched structure, the paired Prentice-Wilcoxon (PPW) test and the Akritas test has been used to assess the treatment effect on paired survival data. They are based on the scores obtained from survival data. However, they might be biased when there exists extreme values in the paired scores. The modi ed PPW and Akritas tests are proposed in the thesis and sensitivity analysis are developed based on the PPW, Akritas, the modi ed PPW and Akritas test. Simulation studies are conducted to compare the performance of these four tests with the weighted logrank test and the marginal structural Cox model. The Akritas and the modi ed Akritas test have the best performance when the hazard ratios are constant and varying over time and when the censoring is dependent. The modi ed tests are applied to a study of primary biliary cirrhosis (PBC) and a sensitivity analysis is conducted.
Bo Lu (Advisor)
Michael Pennell (Committee Member)
89 p.
Biostatistics
propensity score matching; survival; causal inference; sensitivity analysis

Recommended Citations

Citations

  • Zhu, D. (2016). Evaluating causal effect in time-to-event observarional data with propensity score matching [Master's thesis, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1449767366

    APA Style (7th edition)

  • Zhu, Danqi. Evaluating causal effect in time-to-event observarional data with propensity score matching. 2016. Ohio State University, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=osu1449767366.

    MLA Style (8th edition)

  • Zhu, Danqi. "Evaluating causal effect in time-to-event observarional data with propensity score matching." Master's thesis, Ohio State University, 2016. http://rave.ohiolink.edu/etdc/view?acc_num=osu1449767366

    Chicago Manual of Style (17th edition)

osu1449767366
1,921
© 2016, all rights reserved.
This open access ETD is published by The Ohio State University and OhioLINK.