compare: Compare Methods to Divide Steps into "Long" and "Short"

In PCDimension: Finding the Number of Significant Principal Components

Compare Methods to Divide Steps into "Long" and "Short"

Description

Auer and Gervini developed a Bayesian graphical method to determine the number d of significant principal components; a brief overview is included in the help for the AuerGervini class. The output of their method is a step function that displays the maximum a posteriori (MAP) choice of d as a step function of a one-parameter family of prior distributions, and they recommend choosing the highest "long" step. The functions described here help automate the process of dividing the step lengths into "long" and "short" classes.

Usage

  compareAgDimMethods(object, agfuns)

Arguments

object	An object of the AuerGervini class
agfuns	A list of functions

Details

This method simply iterates over the list of functions that implement different algorithms/methods to determine the PC dimension.

Value

Returns an integer vector of te same length as the list of agfuns, containing the number of significant principal components computed by each method.

Author(s)

Kevin R. Coombes <[e-mail]>, Min Wang <[e-mail]>.

References

P Auer, D Gervini. Choosing principal components: a new graphical method based on Bayesian model selection. Communications in Statistics-Simulation and Computation 37 (5), 962-977

See Also

AuerGervini, agDimension.

Examples

# simulate variances
lambda <- rev(sort(diff(sort(c(0, 1, runif(9))))))
# apply the Auer-Gervini method
ag <- AuerGervini(lambda, dd=c(3,10))
# try different methods
agfuns <- list(twice=agDimTwiceMean,
               km=agDimKmeans,
               cpt=agDimCPT)
compareAgDimMethods(ag, agfuns)

PCDimension documentation built on Feb. 12, 2024, 3:01 a.m.