Premium
Exploratory analysis of three‐way data by simultaneous latent budget model
Author(s) -
Tambrea Narcisa,
Siciliano Roberta
Publication year - 1999
Publication title -
applied stochastic models in business and industry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.413
H-Index - 40
eISSN - 1526-4025
pISSN - 1524-1904
DOI - 10.1002/(sici)1526-4025(199910/12)15:4<469::aid-asmb410>3.0.co;2-o
Subject(s) - computer science , rank (graph theory) , set (abstract data type) , data set , exploratory analysis , latent variable , table (database) , latent variable model , econometrics , mixture model , constant (computer programming) , decomposition , statistics , data mining , mathematics , combinatorics , ecology , data science , biology , programming language
In the analysis of a set of tables with constant row sum data the simultaneous latent budget model provides a reduced‐rank decomposition of the observed budgets of each table by a mixture of unknown, or latent, budgets. A constrained least‐squares algorithm is introduced to estimate the model parameters under no probability distribution assumptions. Some examples of applications on real data sets are finally shown. Copyright © 1999 John Wiley & Sons, Ltd.