Detailed information available here

---

---

This workshop will provide descriptions for basic and emerging nonparametric statistical methods and examples for their application.  One emerging method for randomized clinical trials is nonparametric analysis of covariance.  It uses weighted least squares to adjust the difference between two treatments for the means of a response variable to a structure that has no difference for the means of covariables (on the basis of the known equality of means for covariables that randomization of treatments to patients implies).  This method enables variance reduction for estimates of differences between treatments for a response variable without issues concerning assumptions of linearity or parallelism in relationships to covariables.  Another emerging method uses multivariate sets of Mann-Whitney estimates for the probability of better response for one group than another to evaluate relationships pertaining to more than two groups and/or more than one response for the comparison between two groups.  Attention is given to the estimation of the covariance structure for these Mann-Whitney estimates and the use of weighted least squares to fit models to describe their variation.  Many examples will be provided to illustrate the application of all of these nonparametric statistical methods, and there will be some related discussion of computing strategies.

---

Registration form, schedule, map, and other information available here.