Abstract
In medical research clinical trials, patients are usually evaluated on multiple ordinal-scaled psychometric instruments. Conventional treatment efficacy assessment is conducted on each instrument individually. In this presentation, we consider a dose-response study in neuroscience and propose to assess the joint efficacy of the instruments based on a multivariate method which is applied to each stratum of the multivariate baseline measures. The performance of the proposed method will be compared with some conventional approaches. The significance of the proposed method will be evaluated using both the asymptotic distribution of the test statistic and the empirical permutation distribution of the data.
Original language | American English |
---|---|
Title of host publication | Proceedings of the Joint Statistical Meetings |
State | Published - Aug 2 2007 |
Keywords
- Dose Response
- Multivariate Test
- Neuroscience Clinical Trial
- Ordinal Data
DC Disciplines
- Biostatistics
- Community Health
- Public Health