An extension of MLDA to Three-way Contingency Tables
[Une extension de l’analyse discriminante multiblocs aux tableaux de contingence ternaires]
Journal de la société française de statistique, Tome 160 (2019) no. 2, pp. 67-82.

L’objet de cet article est de proposer une extension de l’analyse factorielle discriminante de tableaux multiples à la description d’un ensemble de tableaux de contingence qui ont été observés à différentes occasions et qui ont le même nombre de lignes et le même nombre de colonnes. Cette méthode, MLDA-TCT, est un compromis entre l’analyse factorielle des correspondances et l’analyse discriminante linéaire. MLDA-TCT détermine une ou plusieurs variables auxiliaires pour chaque tableau de données, de telle manière que ces variables prennent en compte à la fois les relations entre les lignes et les colonnes des tableaux de contingence et les relations entre les tableaux de contingence.

The aim of this paper is to propose an extension of Multiblock Linear Discriminant Analysis (MLDA) for analyzing a set of contingency tables which have been observed in different occasions and have the same number of rows and the same number of columns. This extension, Multiblock Linear Discriminant Analysis of Three-way Contingency Tables (MLDA-TCT, is midway between correspondence analysis and linear discriminant analysis; MLDA-TCT computes one or several variables for each data table, such that these variables take into account relationships between rows and columns of the contingency tables in one hand, and in the other hand, take into account relationships between contingency tables.

Keywords: linear discriminant analysis, canonical analysis, correspondence analysis, bi-partitioned data table, Three-way contingency table
Mot clés : analyse discriminante, analyse canonique, analyse des correspondances, tableau bi-partitionné, tableau de contingence ternaire 
@article{JSFS_2019__160_2_67_0,
     author = {Casin, Philippe},
     title = {An extension of {MLDA} to {Three-way} {Contingency} {Tables}},
     journal = {Journal de la soci\'et\'e fran\c{c}aise de statistique},
     pages = {67--82},
     publisher = {Soci\'et\'e fran\c{c}aise de statistique},
     volume = {160},
     number = {2},
     year = {2019},
     mrnumber = {3987790},
     zbl = {1420.62258},
     language = {en},
     url = {http://archive.numdam.org/item/JSFS_2019__160_2_67_0/}
}
TY  - JOUR
AU  - Casin, Philippe
TI  - An extension of MLDA to Three-way Contingency Tables
JO  - Journal de la société française de statistique
PY  - 2019
SP  - 67
EP  - 82
VL  - 160
IS  - 2
PB  - Société française de statistique
UR  - http://archive.numdam.org/item/JSFS_2019__160_2_67_0/
LA  - en
ID  - JSFS_2019__160_2_67_0
ER  - 
%0 Journal Article
%A Casin, Philippe
%T An extension of MLDA to Three-way Contingency Tables
%J Journal de la société française de statistique
%D 2019
%P 67-82
%V 160
%N 2
%I Société française de statistique
%U http://archive.numdam.org/item/JSFS_2019__160_2_67_0/
%G en
%F JSFS_2019__160_2_67_0
Casin, Philippe. An extension of MLDA to Three-way Contingency Tables. Journal de la société française de statistique, Tome 160 (2019) no. 2, pp. 67-82. http://archive.numdam.org/item/JSFS_2019__160_2_67_0/

[1] Aktas, S. Subsymmetry and asymmetry models for multiway square contingency tables with ordered categories, Open mathematics research article, Volume 14 (2016), pp. 195-204 | MR | Zbl

[2] Bougeard, S.; Abdi, H.; Saporta, G.; Niang, N. Clusterwise analysis for multiblock component methods, Advances in Data Analysis and Classificationses Classif (2017), pp. 1-29 | MR | Zbl

[3] Black, W. C.; Babin, B.J.; Anderson, R. E.; Tatham, R.L. Multivariate data analysis, Volume 5 (1998), pp. 207-219

[4] Bécue-Bertaut, M.; Pages, J. A principal axes method for comparing multiple contingency tables: MFACT, Computational Statistics and Data Analysis, Volume 45 (2004), pp. 481-503 | MR | Zbl

[5] Benzécri, J-P. L’analyse des données, Dunod, 1980 | MR | Zbl

[6] Beh, J.; Lombardo, R. Symetrical and non-symetrical three-way correspondance analysis, Correspondance Analysis: Theory, Pratice and New Strategies (2014), pp. 481-503

[7] Beh, J.; Lombardo, R.; Gianmarco, A. Correspondence analysis and the Freeman-Tukey statistic: A study of archaeological data, Computational Statistics & Data Analysis, Volume 128 (2018), pp. 73 -86 | MR | Zbl

[8] Bougeard, S.; Niang, S.; Verron, T.; Bry, X. Current multiblock methods: Competition or complementarity? A comparative study in a unified framework, Chemometrics and Intelligent Laboratory Systems, Volume 182 (2018), pp. 131 -148

[9] Bougeard, S.; Qannari, E.M.; Rose, N. Multiblock Redudancy Analysis: interpretation tools and application in epidemiology, Journal of chemiometrics, Volume 25 (2011), pp. 467-475

[10] Carroll, J-D. Generalization of canonical correlation analysis to three or more sets of variables., Proceedings of the 76th annual convention of the Americain Psychological Association, Volume 3 (1968), pp. 227-228

[11] Casin, Ph. L’analyse factorielle discriminante de tableaux multiples, Journal de la Société Française de Statistique, Volume 156 (2015), pp. 1-20 | Numdam | Zbl

[12] Casin, Ph. Categorical multiblock linear discriminant analysis, Journal of Applied Statistics (2017), pp. 1-14

[13] Cazes, P. L’analyse de certains tableaux rectangulaires décomposés en blocs : généralisation des propriétés rencontrées dans l’analyse des correspondances multiples.IV.Cas modèles, Les cahiers de l’analyse des données, Volume VI(2) (1981), pp. 135-143

[14] Escofier, B.; Pagès, J. Multiple factor analysis (AFMULT package), Computational Statistics and Data Analysis, Volume 18.1994 (1994), pp. 121-140 | Zbl

[15] Eslami, A.; Qannari, E.M.; Kohler, A.; Bougeard, S. Multivariate analysis of multiblock and multigroup data, Chemometrics and Intelligent Laboratory Systems, Volume 133 (2014), pp. 63-69

[16] Fisher, R-A. The use of multiple measurements in taxonomics problems, Annals of eugenics, Volume 7-(2) (1936), pp. 179-188

[17] Hotelling, H. Relations between two sets of variants, Biometrika, Volume 28 (1936), pp. 321-337 | JFM | Zbl

[18] Israels, A. Eigenvalues techniques for qualitative data, DSWO Press, 1987

[19] Jollife, I-T. Principal Component Analysis, Springer, 2002

[20] Kostov, B.; Bécue-Bertaut, M.; Husson, F. Multiple Factor Analysis for Contingency Tables in the FactoMineR Package, The R Journal, Volume 5 (2013), pp. 29-38

[21] Kang, M.; Kim, D-C.; Liu, C.; Gao, J. Multiblock Discriminant Analysis for Integrative Genomic Study, Biomed Research International, Volume 2015 (2015), pp. 1-10

[22] Kroonenberg, P.; Lombardo, R. Nonsymmetric Correspondence Analysis: A Tool for Analysing Contingency Tables with a Dependence Structure, Multivariate Behavioral Research, Volume 34 (1999) no. 3, pp. 367-396

[23] Kateri, M.; Petros Dellaportas, P. Conditional symmetry models for three-way contingency tables, Journal of Statistical Planning and Inference, Volume 142 (2012) no. 8, pp. 2430 -2439 | Zbl

[24] Lavit, C. Analyse conjointe de tableaux quantitatifs, Masson, Paris, 1988

[25] Leclerc, A. L’analyse factorielle des correspondances sur juxtaposition de tableaux de contingence, Revue de statistique appliquée, Volume 23-3 (1975), pp. 5-16

[26] Lombardo, R. Three-way association measure decompositions: The Delta index, Journal of Statistical Planning and Inference, Volume 141 (2011) no. 5, pp. 1789 -1799 | Zbl

[27] Saporta, G. Liaison entre plusieurs ensembles de variables et codage de variables qualitatives, Thèse, Université de Paris VI. (1976)

[28] Sabatier, R.; Vivien, M.; Reynès, C. Une nouvelle proposition, l’analyse discriminante Multitableaux : STATIS-LDA, Journal de la Société Française de Statistique, Volume 154 (2013), pp. 31-43 | Numdam | Zbl

[29] Taneichi, Nobuhiro; Sekiya, Yuri; Toyama, Jun Transformed statistics for tests of conditional independence in J x K x L contingency tables, Journal of Multivariate Analysis, Volume 171 (2019), pp. 193 -208 | Zbl

[30] Tenenhaus, A.; Tenenhaus, M. Regularized generalizd canonical correlation analysis for multiblock and multigroup data analysis, Journal of operational research, Volume 238 (2014), pp. 391-403 | Zbl

[31] Vallejo-Arnadela, A.; Vincente-Villardon, J.; Gamindo-Villardon, M. Canonical-STATIS : Biplot analysis of multi-group structured data based on STATIS-act methodology, Computational Statistics and Data Analysis, Volume 46 (2007), pp. 4193-4205

[32] Xiang, F.; Nie, G.; Meng, G; Pan, C.; Zhang, C. Discriminant least squares regression for multiclass classification and feature selection, IEE transactions on neural networks and learning systems, Volume 23 (2012), pp. 1738-1754

[33] Zarraga, A.; Goitisolo, B. Méthode factorielle pour l’analyse simultanée de tableaux de contingence, Revue de statistique appliquée, Volume 50(2) (2002), pp. 47-70

[34] Zarraga, A.; Goitisolo, B. Etude de la structure inter-tableaux a travers l’analyse simultanée, Revue de statistique appliquée, Volume 51 (3) (2003), pp. 117-142

[35] Zarraga, A.; Goitisolo, B. Simultaneous analysis and multiple factor analysis for contingency tables: Two methods for the joint study of contingency tables, Computational Statistics & Data Analysis, Volume 53 (2009), pp. 3171 -3182 | Zbl