B1463
Title: New construction of a cylindrical distribution from two base distributions
Authors: Tomoaki Imoto - University of Shizuoka (Japan) [presenting]
Abstract: In diverse scientific fields, data often appear that can be represented as points in the circumference of a unit circle, called circular data. In many cases, observations are made on linear and circular variables to discover some relationships like wind speed and direction. Such data is called cylindrical data. We propose a method for constructing a distribution for modeling cylindrical data. The difference from previous work is that the support of the linear variable of the constructed distribution can be arbitrary. This method is one of the specific marginals methods, and the conditional distribution of the circular variable becomes a well-studied distribution belonging to a family of sine-skewed distributions. The other research about estimation and application for fitting real data are also shown.
