J. L. Vicente Villardón, L. Vicente-González
Classical Biplot methods allow the simultaneous representation of individuals and variables of a data matrix. Recently we have proposed a biplot representation based on logistic response models for binary data. The coordinates of individuals and variables are computed to have logistic responses along the biplot dimensions. The method is related to logistic regression in the same way as Classical Biplots are related to linear regression and we refer to the method as Logistic Biplot. In the same way as Linear Biplots are related to Principal Components Analysis, Logistic Biplots are related to Latent Trait Analysis. The estimation procedures are based on joint or marginal maximum likelihood suitable for small data matrices.
In this paper we study some alternative algorithms for parameter estimation, based on gradient descent, that can be applied to bigger data sets.
The new procedures are illustrated using data on SNPs (Single Nucleotide Polymorphisms) from the HAPMAP project.
Palabras clave / Keywords: logistic biplot, binary data
Programado
Sesión GT02-2: Análisis Multivariante y Clasificación (AMyC-2). Organizadora: Eva Boj del Val
29 de mayo de 2018 12:20
Sala 5
