n this paper we present an unification to the methods for graphically summarising the association between two categorical variables that form a two-way contingency table. In particular we focus on following methods based on the decomposition of a known index: correspondence analysis (CA) based on the decomposition of Pearson’s chi-squared statistic; non symmetrical correspondence analysis (NSCA) based on the decomposition of the Goodman-Kruskal tau index; singly ordered cumulative correspondence analysis based on the decomposition of Taguchi’s statistic; doubly ordered cumulative correspondence analysis based on the decomposition of the doubly cumulative chi-squared statistic.
An unified approach to simple correspondence analysis
CAMMINATIELLO, Ida;
2012
Abstract
n this paper we present an unification to the methods for graphically summarising the association between two categorical variables that form a two-way contingency table. In particular we focus on following methods based on the decomposition of a known index: correspondence analysis (CA) based on the decomposition of Pearson’s chi-squared statistic; non symmetrical correspondence analysis (NSCA) based on the decomposition of the Goodman-Kruskal tau index; singly ordered cumulative correspondence analysis based on the decomposition of Taguchi’s statistic; doubly ordered cumulative correspondence analysis based on the decomposition of the doubly cumulative chi-squared statistic.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
