Abstract
Suppose that K ≥ 2 treatments are being compared in a clinical trial. If complete randomization is used, the next patient is equally likely to be assigned to any of the K treatments, so that the treatment allocation probabilities
are all 1/K. So this randomization rule does not take into account the previous treatment assignments, responses and covariate vectors, and the current patient’s covariate vector.
are all 1/K. So this randomization rule does not take into account the previous treatment assignments, responses and covariate vectors, and the current patient’s covariate vector.
| Original language | English |
|---|---|
| Title of host publication | Modern Adaptive Randomized Clinical Trials |
| Subtitle of host publication | Statistical and Practical Aspects |
| Editors | Oleksandr Sverdlov |
| Place of Publication | New York |
| Publisher | Chapman and Hall/CRC |
| Chapter | 7 |
| Pages | 159-175 |
| Number of pages | 16 |
| ISBN (Electronic) | 9780429161124 |
| ISBN (Print) | 9781482239881 |
| Publication status | Published - Jun 2015 |
Fingerprint
Dive into the research topics of 'Statistical inference following covariate-adaptive randomization: Recent advances'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver