mpCOVSTATIS: mpCOVSTATIS: Core Function for COVSTATIS via MExPosition
In MExPosition: Multi-table ExPosition

Description Usage Arguments Details Value Author(s) References See Also Examples

All COVSTATIS steps are combined in this function. It enables preparation of the data, processing and graphing.

1 2	mpCOVSTATIS(data, normalization = 'None', masses = NULL, table = NULL, make.table.nominal = TRUE, DESIGN = NULL, make.design.nominal = TRUE, graphs = TRUE)

`data`	Matrix of preprocessed data
`normalization`	String option of either 'None', 'MFA' (DEFAULT), or 'Sum_PCA'
`masses`	Masses
`table`	Design Matrix - used to identifty the tables of the data matrix
`make.table.nominal`	a boolean. If TRUE (default), table is a vector that indicates tables (and will be dummy-coded). If FALSE, table is a dummy-coded matrix.
`DESIGN`	a design matrix to indicate if rows belong to groups.
`make.design.nominal`	Boolean option. If TRUE (default), table is a vector that indicates groups (and will be dummy-coded). If FALSE, table is a dummy-coded matrix.
`graphs`	Boolean option. If TRUE (default), graphs are displayed

COVSTATIS is used to analysis covariance matrices. It is an extension of three-way multidimensional scaling.

Returns a large list of items which are divided into four categories:

`$Overview`	Overview of Results
`$InnerProduct`	Results for the Inner Product
`$Compromise`	Results for the Compromise
`$Table`	Results for the Tables

The results for Overview are bundled inside of $Overview.

`$Overview$data`	Data Matrix
`$Overview$normalization`	Type of normalization used
`$Overview$table`	Matrix used to identify the different tables of the data matrix
`$Overview$num.groups`	Number of Tables

The results for InnerProduct are bundled inside of $InnerProduct

`$InnerProduct$S`	Inner Product: Scalar Product Matrices
`$InnerProduct$C`	Inner Product: C Matrix
`$InnerProduct$rvMatrix`	Inner Product: RV Matrix
`$InnerProduct$eigs.vector`	Inner Product: Eigen Vectors
`$InnerProduct$eigs`	Inner Product: Eigen Values
`$InnerProduct$fi`	Inner Product: Factor Scores
`$InnerProduct$t`	Inner Product: Percent Variance Explained
`$InnerProduct$ci`	Inner Product: Contribution of the Rows
`$InnerProduct$cj`	Inner Product: Contribution of the Columns
`$InnerProduct$alphaWeights`	Alpha Weights

The results for the Compromise are bundled inside of $Compromise

`compromise`	Compromise Matrix
`compromise.eigs`	Compromise: Eigen Values
`compromise.eigs.vector`	Compromise: Eigen Vector
`compromise.fi`	Compromise: Factor Scores
`Compromise.t`	Compromise: Percent Variance Explained
`compromise.ci`	Compromise: Contributions of the rows
`compromise.cj`	Compromise: Contributions of the Columns

The results for the Tables are bundled inside of $Table.

`$m`	Table: masses
`$Table$eigs`	Table: Eigen Values
`$Table$eigs.vector`	Table: Eigen Vectors
`$Table$Q`	Table: Loadings
`$Table$fi`	Table: Factor Scores
`$Table$partial.fi`	Table: Partial Factor Scores
`$Table$partial.fi.array`	Table: Arrray of Partial Factor Scores
`Table$ci`	Table: Contribition of the Rows
`$Table$cj`	Table: Contribution of the Columns
`$Table$t`	Table: Percent Variance Explained

Cherise R. Chin Fatt and Hervé Abdi.

Abdi, H., Williams, L.J., Valentin, D., & Bennani-Dosse, M. (2012). STATIS and DISTATIS: Optimum multi-table principal component analysis and three way metric multidimensional scaling. Wiley Interdisciplinary Reviews: Computational Statistics, 4, 124-167

mpCANOSTATIS

  #COVSTATIS
   data('faces2005')
   table = c('pixel','pixel','pixel','pixel','pixel','pixel', 
   'distance','distance','distance','distance','distance','distance', 
   'ratings','ratings','ratings','ratings','ratings','ratings',
   'similarity','similarity','similarity','similarity','similarity','similarity')
   demo.covstatis.2005 <- mpCOVSTATIS(faces2005$data, table = table)