A0489
Title: Covariance estimation for histograms using copulas
Authors: Lynne Billard - University of Georgia (United States) [presenting]
Abstract: Histogram-valued data are emerging increasingly often as a consequence of the aggregation of large data sets. One statistic that underpins many methodologies, especially regression and principal component analyses, is the covariance function. Typically, unfortunately, only the marginal distributions are recorded. Therefore, maximum likelihood, inference function for margins, and canonical maximum likelihood estimation methods based on copula functions are proposed. These are then used to obtain estimators of the underlying covariances.
