lakhesize: Lakhesize

In lakhesis: Consensus Seriation for Binary Data

Lakhesize

Description

This function returns the row and column consensus seriation for a list of strands, containing their rankings, the results of their PCA, and coefficients of association and concentration.

Usage

lakhesize(strands, obj)

Arguments

strands	A list of strands, which are data frames returned by ca.procrustes.curve.
obj	The intial incidence matrix.

Details

Consensus seriation is achieved by iterative, multi-step linear regression using simulation. On one iteration, strands are chosen at random, omitting incomplete or missing pairs, using PCA to determine the best-fitting line for their rankings. Both strands' rankings are then regressed onto that line to determine missing values, and then re-ranked, repeating until all strands have been regressed. PCA of the simulated rankings is then used to determine the final sequence of the row and column elements.

Value

A list of the following:

• RowConsensus Data frame of the consensus seriation of the row elements in the order of their projection on the first principal axis. Contains one column, Row.

• ColConsensus Data frame of the consensus seriation of the column elements in the order of their project onto the first principal axis. Contains one column, Column.

• RowPCA The results of ⁠\link[stats]{prcomp}⁠ performed on the row elements of strands.

• ColPCA The results of ⁠\link[stats]{prcomp}⁠ performed on the column elements of strands.

• Coef A data frame containing the coefficients of agreement and concentration:

  • Strand The number of the strand.

  • Consensus.Spearman.Sq the measure of agreement, i.e., how well each strand accords with the consensus seriation. Using the square of Spearman's rank correlation coefficient, \rho^2, between each strand and the consensus ranking, agreement is computed as the product of \rho^2 for their row and column rankings, \rho_r^2\rho_c^2.

  • Concentration.Kappa the concentration coefficient \kappa, which provides a measure of the optimality of each strand (see kappa.coef).

Examples

data("quattrofontanili")
data("qfStrands")
lakhesize(qfStrands, quattrofontanili)

lakhesis documentation built on June 22, 2024, 10:27 a.m.