A0464
Title: A simplification of aggregated symbolic data
Authors: Junji Nakano - Chuo University (Japan) [presenting]
Nobuo Shimizu - The Institute of Statistical Mathematics (Japan)
Yoshikazu Yamamoto - Tokushima Bunri University (Japan)
Abstract: The interest is in comparing groups of individuals, where each individual is described by observations of continuous and categorical variables. To summarize each group, the number of individuals, the first and second moments of continuous variables and dummy variables for categorical variables are used. Such statistics are called aggregated symbolic data (ASD). Although ASD is an appropriate summary of a group, it is still complicated. There is a need to simplify it more for intuitive understanding and visualization. The aim is to treat continuous variables and categorical variables equally in the simplification by defining appropriate scores for categorical values. The method of multiple correspondence analysis is used to determine scores for categorical values. A visualization of the simplified ASD is also presented.
