Update the Lambda parameters of clusters.

Share:

Description

Updates the Lambda parameters to maximize the likelihood of the data under Mallows' model.

Usage

1
2
  UpdateLambda(r, R, z, G, dists.to.Rg, dists.table,
    top.bound = 1000)

Arguments

r

Matrix of partial rankings.

R

Current modal sequences.

z

Current probabilities of memberships in each cluster.

G

Number of modal sequences.

dists.to.Rg

Matrix of the distances between the data and the current modal sequences.

dists.table

Table of the distance distribution in N! space, under Kendall's metric.

top.bound

The maximum value for the lambda parameter.

Value

Vector of new lambda parameters for the clusters.

Author(s)

Erik Gregory

References

"Mixtures of distance-based models for ranking data". Thomas Brendan Murphy & Donal Martin. 1 April 2002. Computational Statistics & Data Analysis 41 (2003) 645-655.

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.