B1030
Title: Minimum information dependence modeling
Authors: Keisuke Yano - The Institute of Statistical Mathematics (Japan) [presenting]
Tomonari Sei - The University of Tokyo (Japan)
Abstract: A method of dependence modeling for a broad class of multivariate data is proposed. Our class is characterized by two orthogonal sets of parameters: the parameters of dependence and those of marginal distributions. We present the existence and uniqueness theorem for our model. To estimate the dependence parameter, we establish conditional inference together with a sampling procedure and show that conditional inference is asymptotically indistinguishable from the maximum likelihood inference. We also discuss the information-geometrical structure and the connection to the entropic optimal transport and the Schrodinger bridge problems. Finally, we illustrate an application to the earthquake data.
