PLACE:  319 Snedecor

SPEAKER:
Heike Hofman
Augsburg University, Augsburg, Germany

TITLE:
Graphical Methods for the Exploration of Multivariate Categorical Data

ABSTRACT:

In contrast to quantitative data, graphical methods for categorical data are
not well developed.

A scatterplot of two continuous variables Y vs. X, for example, enables us to
draw (rough) conclusions on the correlation coefficient between X and Y and
gives information about how well a regression will fit the data.

For categorical data graphical displays, also, exist, even for high-dimensional
situations.  But the connection between the graphical display and the model is
far less explored for categorical data than for continuous data.

The overall objective of the talk is to give an introduction to the
possibilities of analyzing categorical data by means of graphical tools.

Mosaicplots (Hartigan & Kleiner 1981), here, provide a tool for
multi-dimensional visualization.  Enhanced with interactive features such as
querying or reordering facilities of variables and groups these plots become a
very powerful and easy to use tool for analyzing and understanding multivariate
contexts.
 
 

COFFEE: 3:45 p.m., 104 Snedecor