Research interests

​

• Order restricted inference has been my main area of research for the past ten years. I have a special interest in the interface between the theory of stochastic order relations and non and semi parametric statistical inference for univariate and multivariate outcomes.

​

• Methods and models for ranking and rating: I am interested mainly in models for paired comparison data with and without covariates. I am currently developing a class of models referred to as graph GLMs and studying ways of incorporating constraints into ranking problems. 
   

• Nonparametric methods for multivariate and high dimensional data: in particular I have interest in robust rank based methods for multivariate and high dimensional data, these methodologies are designed to replace classic methods with robust estimation, testing and classification methods that can scale to deal with high dimensional data.
   

• Biostatistics and general statistical methodology: especially case control studies, early detection (screening) programs and ROC methodology. Additional areas of interest are biased sampling, double sampling, and  compositional data analysis.