Package edu.cmu.tetrad.search

Class Gffc

java.lang.Object

edu.cmu.tetrad.search.Gffc

public class Gffc
extends Object

Generalized Find Factor Clusters (GFFC). This generalized FOFC and FTFC to first find clusters using pure 2-tads (pure tetrads) and then clusters using pure 3-tads (pure sextads) out of the remaining variables. We do not use an n-tad test here since we need to check rank, so we will check rank directly. (This is equivqalent to using the CCA n-tad test.)

Kummerfeld, E., & Ramsey, J. (2016, August). Causal clustering for 1-factor measurement models. In Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining (pp. 1655-1664).

Spirtes, P., Glymour, C. N., Scheines, R., & Heckerman, D. (2000). Causation, prediction, and search. MIT press.

Version:

$Id: $Id

Author:

erichkummerfeld, peterspirtes, josephramsey

See Also:

• Ftfc

Constructor Summary

Constructors

Constructor	Description
Gffc(DataSet dataSet, double alpha, int rMax, int ess)	Constructs a new Gffc instance with the specified data set, alpha value, maximum rank (rMax), and effective sample size (ess).

Method Summary

Modifier and Type	Method	Description
Map<List<Integer>,Integer>	findClusters()	Runs the search and returns a graph of clusters with the ir respective latent parents.
void	setVerbose(boolean verbose)	Enables or disables verbose mode for the current instance and associated components.

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

Gffc

public Gffc(DataSet dataSet,
 double alpha,
 int rMax,
 int ess)

Constructs a new Gffc instance with the specified data set, alpha value, maximum rank (rMax), and effective sample size (ess). This implementation initializes several internal structures used for clustering and correlation computations.

Parameters:

dataSet - The data set containing the variables and data for analysis.

alpha - The significance level for statistical tests, typically a small value like 0.05.

rMax - The maximum rank for search operations, influencing the depth and computational expense.

ess - The effective sample size, used for statistical adjustments in computations.

Method Details

findClusters

public Map<List<Integer>,Integer> findClusters()

Runs the search and returns a graph of clusters with the ir respective latent parents.

Returns:

This a map from discovered clusters to their ranks.

setVerbose

public void setVerbose(boolean verbose)

Enables or disables verbose mode for the current instance and associated components.

Parameters:

verbose - a boolean value where true enables verbose mode and false disables it.