Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
416299 | Computational Statistics & Data Analysis | 2006 | 21 Pages |
Abstract
The geometrical interpretation of the INDSCAL model is described in detail. This geometrical interpretation offers new lights about the characteristics of the model which are important for applications. Mainly, correlation between dimensions, range of subject weights, interpretation of subject weights as relationship measurements, role of the mean configuration. Moreover, new helps are proposed for the interpretation and the link between INDSCAL model and multiple factor analysis (MFA) is established. These two methods bring complementary results, MFA being the exploratory approach corresponding to the INDSCAL model. These results are illustrated using a large data set.
Keywords
Related Topics
Physical Sciences and Engineering
Computer Science
Computational Theory and Mathematics
Authors
F. Husson, J. Pagès,