ANalysis Of VAriance (ANOVA) is a method to decompose the total variation of the observations into sum of variations due to different factors and the residual component. When the data are nominal, the usual approach of considering the total variation in response variable as measure of dispersion about the mean is not well defined. Light and Margolin (1971) proposed CATegorical ANalysis Of VAriance (CATANOVA), to analyze the categorical data. Onukogu (1985) extended the CATANOVA method to two-way classified nominal data. The components (sums of squares) are, however, not orthogonal. Singh (1996) developed a CATANOVA procedure that gives orthogonal sums of squares and defined test statistics and their asymptotic null distributions. In order to study which exploratory categories are influential factors for the response variable we propose to apply Non-Symmetrical Correspondence Analysis (D'Ambra and Lauro, 1989) on significant components. Finally, we illustrate the analysis numerically, with a practical example.
Visualization of the Significant Explicative Categories using Catanova Method and Non-Symmetrical Correspondence Analysis for Evaluation of Passenger Satisfaction / I., Camminatiello; D'Ambra, Luigi. - In: JOURNAL OF APPLIED QUANTITATIVE METHODS. - ISSN 1842-4562. - ELETTRONICO. - 5:1(2010), pp. 64-72.
Visualization of the Significant Explicative Categories using Catanova Method and Non-Symmetrical Correspondence Analysis for Evaluation of Passenger Satisfaction
D'AMBRA, LUIGI
2010
Abstract
ANalysis Of VAriance (ANOVA) is a method to decompose the total variation of the observations into sum of variations due to different factors and the residual component. When the data are nominal, the usual approach of considering the total variation in response variable as measure of dispersion about the mean is not well defined. Light and Margolin (1971) proposed CATegorical ANalysis Of VAriance (CATANOVA), to analyze the categorical data. Onukogu (1985) extended the CATANOVA method to two-way classified nominal data. The components (sums of squares) are, however, not orthogonal. Singh (1996) developed a CATANOVA procedure that gives orthogonal sums of squares and defined test statistics and their asymptotic null distributions. In order to study which exploratory categories are influential factors for the response variable we propose to apply Non-Symmetrical Correspondence Analysis (D'Ambra and Lauro, 1989) on significant components. Finally, we illustrate the analysis numerically, with a practical example.File | Dimensione | Formato | |
---|---|---|---|
camminatiello_ambraComp.pdf
accesso aperto
Tipologia:
Documento in Post-print
Licenza:
Creative commons
Dimensione
335.98 kB
Formato
Adobe PDF
|
335.98 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.