Sample Size Re-Estimation: Nonparametric Approach
A question of importance that arises in clinical trials pertains to the number of additional observations, if any, that are required beyond those initially planned. This has been answered in the case of two-treatment double blind clinical experiment via a nonparametric approach in which the mean and the variance of the underlying populations are not functionally related. In this paper, a totally nonparametric approach is taken. We will be interested in testing the equality of two treatment effects against the alternative that one effect is stochastically larger than the other. Using Mann-Whitney-Wilcoxon test criterion, the robustness of the sample size in terms of the level of significance and power of a specified nonparametric alternative is established. This procedure is applied to the data pertaining to blood cholesterol levels in samples drawn from two different age groups.
AMS Subject Classification62G10
KeywordsRobustness Clinical trials Double-blind experiment Sample size re-estimation Arbitrary populations
Unable to display preview. Download preview PDF.
- Govindarajulu, Z., 2002. Robustness of sample size re-estimation procedure in clinical trials, Advances in Methodological and Applied Aspects of Probability and Statistics, N. Balakrishnan (Ed.), Taylor & Francis, New York, 383–398.Google Scholar
- Shih, W.J., 1992. Sample size re-estimation in clinical trials, Biopharmaceutical Sequential Applications (Ed. K.E. Peace) Marcel Dekker Inc., New York, 285–301.Google Scholar