Sequential testing is used in clinical trials and online experiments to terminate trials early if there is sufficient evidence of a treatment effect, reducing the subjects’ exposure to potentially harmful treatments. However, incomplete follow-up of trial subjects can lead to biased estimation of the treatment effect. This thesis aims to evaluate the method proposed by Kittelson et al. (2005) for utilizing measurements from subjects with incomplete follow-up. The evaluation compares the proposed method with three other methods based on size-adjusted power and stopping time, using a Monte Carlo study with varying treatment effects and time point measurement correlations. The proposed method performs better in terms of size-adjusted power and can detect differences between treatment and control groups at an earlier stage than the other methods. The false positive rate observed using the proposed method is also relatively close to the intended false positive rate in all cases.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:uu-505473 |
| Date | January 2023 |
| Creators | Lanzén, Emil |
| Publisher | Uppsala universitet, Statistiska institutionen |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0016 seconds