Functional Data with Spatially Varying Shape Parameters

• Activity Number: 661
• Type: Topic Contributed
• Date/Time: Thursday, August 5, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Section on Nonparametric Statistics
• Author(s): Ana-Maria Staicu*+ and David Ruppert and Ciprian Crainiceanu
• Companies: North Carolina State University and Cornell University and The Johns Hopkins University
• Keywords: Functional Data ; Brain tractography ; Quantile model

Abstract

We propose a novel methodological framework for the analysis of functional data that display spatial heterogeneous skewness. Such data are not envisaged by the current approaches to model functional data, due to the lack of Gaussian - like features. Our methodology allows modeling the pointwise quantiles, has interpretability advantages and is computationally feasible. The methods are inspired by and applied to data obtained from a state-of-the-art brain tractography study. However our models are general and will be relevant to data sets where the object of inference are functions, which have distributional shape characteristics that vary across point locations.