clustord-package: clustord: Clustering Using Proportional Odds Model, Ordered...

In clustord: Cluster Ordinal Data via Proportional Odds or Ordered Stereotype

clustord: Clustering Using Proportional Odds Model, Ordered Stereotype Model or Binary Model.

Description

Biclustering, row clustering and column clustering using the proportional odds model (POM), ordered stereotype model (OSM) or binary model for ordinal categorical data.

Details

The clustord package provides six functions: clustord(), rerun(), mat2df(), calc.SE.rowcluster(), calc.SE.bicluster(), and calc.cluster.comparisons().

Clustering function

The main function is clustord(), which fits a clustering model to the data. The model is fitted using likelihood-based clustering via the EM algorithm. The package assumes that you started with a data matrix of responses, though you will need to convert that data matrix into a long-form data frame before running clustord. Every element in the original data matrix becomes one row in the data frame, and the row and column indices from the data matrix become the columns ROW and COL in the data frame. You can perform clustering on rows or columns of the data matrix, or biclustering on both rows and columns simultaneously. You can include any number of covariates for rows and covariates for columns. Ordinal models used in the package are Ordered Stereotype Model (OSM), Proportional Odds Model (POM) and a dedicated Binary Model for binary data.

The rerun() function is useful for continuing clustering runs that did not converge on the first attempt, and for running new clustering runs using the estimated parameters of a previous run as a starting point. The main input for this function is a clustord object output by clustord, and internally the rerun function runs clustord, after setting up all the input parameters based on the original model fitting run.#'

Utility function

mat2df() is a utility function provided to convert a data matrix of responses into the long-form data frame format required by clustord(), and can also attach any covariates to that long-form data frame if needed.

SE calculation functions

calc.SE.rowcluster() and calc.SE.bicluster() are functions to run after running clustord(), to calculate the standard errors on the parameters fitted using clustord().

Clustering comparisons

calc.cluster.comparisons() can be used to compare the assigned cluster memberships of the rows or columns of the data matrix from two different clustering fits, in a way that avoids the label-switching problem.

Author(s)

Maintainer: Louise McMillan [e-mail] (ORCID) [copyright holder]

Authors:

• Daniel Fernández Martínez [e-mail] (ORCID)

• Ying Cui [e-mail]

• Eleni Matechou [e-mail] (ORCID)

Other contributors:

• W. N. Venables (clustord osm regression functions and S3 methods derived by Louise McMillan from MASS package polr function by Venables and Ripley) [contributor, copyright holder]

• B. D. Ripley (clustord osm regression functions and S3 methods derived by Louise McMillan from MASS package polr function by Venables and Ripley) [contributor, copyright holder]

See Also

Useful links:

• https://vuw-clustering.github.io/clustord/

clustord documentation built on June 8, 2025, 1:03 p.m.