Correspondence analysis in R

From InterSciWiki

Jump to: navigation, search


See Stephen P. Borgatti - Anthropac

Contents

[edit] Background

CA=Wikipedia:Principal components analysis#Correspondence analysis, invented by Benzécri (J.-P. Benzécri et coll., L'Analyse des Données. Tome I, Taxinomie. Tome II, Correspondances. Dunod, 1973 (2nd edn., 1976): "conceptually similar to PCA, but scales the data (which must be positive) so that rows and columns are treated equivalently." As in factor analysis, many variables can be scaled at once, and more compactly (with one fewer factor removed which also removes the distortions of "marginal effects." "It is [also] applied to contingency tables. [Here] Correspondence analysis decomposes the Chi-square statistic associated to this table into orthogonal factors. Because correspondence analysis is a descriptive technique, it can be applied to tables for which the Chi-square statistic is appropriate or not." PCA=Principal components analysis.

Notes on Correspondence analysis from Statistics, North Carolina State

CA and a Factor analysis shortcoming

CA for 2- and 3-way tables

[edit] R software

Correspondence analysis and Data Coding in R and Java by Fionn Murtagh, Chapman & Hall/CRC, 2005, pp 250+xviii.

and here is the R code! also with Supplementary rows and Supplementary columns

There are also, in the web sites with this book, files for

  • Hierarchical clustering
  • Data sets
  • Text processing
  • continuing ground-breaking innovation underlying data coding in the CA context

[edit] Other references

Sten Erik Clausen. Applied Correspondence Analysis: An Introduction. Sage.


Correspondence analysis 1995 Dianne Phillips Social Research Update 7

Multiple Correspondence Analysis 2006 Michael J. Greenacre

[edit] Relocate

Wikipedia:Analysis of categorical data

Personal tools