Year: 2006
Type: book-chapter
Abstract: In analysis of variance, there is usually little attention for interpreting\nthe terms of the effects themselves, especially for interaction\neffects. One of the reasons is that the number of interaction-effect\nterms increases rapidly with the number of predictor variables and\nthe number of categories. In this paper, we propose a new model,\ncalled the interaction decomposition model, that allows to visualize\nthe interactions. We argue that with the help of the visualization, the\ninteraction-effect terms are much easier to interpret. We apply our\nmethod to predict holiday spending1 using seven categorical predictor\nvariables.
Authors: Patrick Groenen, Alex Koning
Institution: Erasmus University Rotterdam
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2139686397