Parallel coordinates for exploratory modelling analysis

Authors:
Antony Unwin;Chris Volinsky;Sylvia Winkler
Affiliations:
Computer-Oriented Statistics & Data Analysis, Mathematics Institute, University of Augsburg, Augsburg 86135, Germany;AT&T Labs Florham Park, New Jersey;Computer-Oriented Statistics & Data Analysis, Mathematics Institute, University of Augsburg, Augsburg 86135, Germany
Venue:
Computational Statistics & Data Analysis - Data visualization
Year:
2003

Citing 1
Cited 2

A method for simultaneous variable selection and outlier identification in linear regression

Computational Statistics & Data Analysis

Discussion: Interacting with parallel coordinates

Interacting with Computers
High-dimensional data visualisation: The textile plot

Computational Statistics & Data Analysis

Quantified Score

Hi-index	0.00

Visualization

Abstract

Parallel coordinate plots have been shown to be useful in exploratory data analysis, especially when they are implemented in interactive software. They are also very useful in exploratory modelling analysis, that is in the evaluation and comparison of many models simultaneously, where they can be applied in at least three quite different ways. The advantages of this innovative graphical tool will be illustrated using the software CASSATT and a data set previously analysed using Bayesian Model Averaging.