DistStatis-Class: Class 'DistStatis' DistStatis S4 class (linkHD:Multiple...

In lauzingaretti/LinkHD: LinkHD: a versatile framework to explore and integrate heterogeneous data

Class DistStatis DistStatis S4 class (linkHD:Multiple Heterogeneous Dataset Integration) Statis with Distance options implementation.

Description

Class DistStatis DistStatis S4 class (linkHD:Multiple Heterogeneous Dataset Integration) Statis with Distance options implementation.

Features

1. DistStatis (implements Statis method incorporating Distance options to integrate multiple heterogeneous datasets)

2. Implement a LM (Linear Model) to variable selection

3. Incorporate a method to variable clustering

4. Incorporate some visualization tools: Compromise visualization, Relationship-visualization

Fields

• RV: Vectorial Correlation Matrix between studies.

• Inertia.RV: Inertia (%) explained for all tables.

• Euclid.Im: Euclidean Image of all studies.

• Inertia.comp: Inertia (%) explained for all dimensions of compromise matrix.

• Compromise.Coords: Projection of all observations in compromise (Coords).

• Compromise.Matrix: Compromise Matrix from statis methodology.

• RQO: Representation Quality of observations in compromise matrix.

• TableProjections: Projection of each table on Compromise configuration

@slot RV: Vectorial Correlation Matrix between studies. @slot Inertia.RV: Inertia (%) explained for all tables. @slot Euclid.Im: Euclidean Image of all studies. @slot Inertia.comp: Inertia (%) explained for all dimensions of compromise matrix. @slot Compromise.Coords: Projection of all observations in compromise (Coords). @slot Compromise.Matrix: Compromise Matrix from statis methodology. @slot RQO: Representation Quality of observations in compromise matrix. @slot TableProjections: Projection of each table on Compromise configuration

DistStatis-general-functions

DistStatis

Getters for their respective slots.

@author Laura M Zingaretti

Examples

{
showClass('DistStatis')
}

lauzingaretti/LinkHD documentation built on March 7, 2023, 9:21 a.m.