Package edu.cmu.tetrad.search.test

Class Kci

java.lang.Object

edu.cmu.tetrad.search.test.Kci

All Implemented Interfaces:

IndependenceTest, RowsSettable

public class Kci
extends Object
implements IndependenceTest, RowsSettable

Gives an implementation of the Kernel Independence Test (KCI) by Kun Zhang, which is a general test of conditional independence. The reference is here:

Zhang, K., Peters, J., Janzing, D., and Schölkopf, B. (2012). Kernel-based conditional independence test and application in causal discovery. arXiv preprint arXiv:1202.3775.

Please see that paper, especially Theorem 4 and Proposition 5.

Using optimal kernel bandwidths suggested by Bowman and Azzalini (1997):

Bowman, A. W., and Azzalini, A. (1997). Applied smoothing techniques for data analysis: the kernel approach with S-Plus illustrations (Vol. 18). OUP Oxford.

Version:

$Id: $Id

Author:

kunzhang, Vineet Raghu on 7/3/2016, josephramsey refactoring 7/4/2018, 12/6/2024

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static final record	Kci.EigenReturn	A record representing the result of an eigenvalue decomposition.
static enum	Kci.KernelType	Represents the type of kernel to be used in a computation.

Constructor Summary

Constructors

Constructor	Description
Kci(DataSet data, double alpha)	Constructor.

Method Summary

Modifier and Type	Method	Description
IndependenceResult	checkIndependence(Node x, Node y, Set<Node> z)	Checks the independence between two nodes given a set of conditioning variables.
static double[]	convertTo1DArray(org.ejml.simple.SimpleMatrix matrix)	Converts a SimpleMatrix to a 1D array.
double	getAlpha()	Returns the significance level of the independence test.
ICovarianceMatrix	getCov()	Returns the covariance matrix.
DataModel	getData()	Returns The data model for the independence test.
List<DataSet>	getDataSets()	Returns a list consisting of the dataset for this test.
List<Integer>	getRows()	Returns the rows used in the test.
int	getSampleSize()	Returns the sample size.
double	getScore(IndependenceResult result)	Returns alpha - p.
Node	getVariable(String name)	Returns the variable of the given name.
List<Node>	getVariables()	Returns the list of variables over which this independence checker is capable of determinining independence relations.
IndependenceTest	indTestSubset(List<Node> vars)	Returns an Independence test for a sublist of the variables.
boolean	isVerbose()	Returns the value of the verbose flag.
void	setAlpha(double alpha)	Sets the alpha level for the test.
void	setApproximate(boolean approximate)	Sets whether the approximate algorithm should be used.
void	setKernelType(Kci.KernelType kernelType)	Sets the type of kernel to be used in computations.
void	setNumBootstraps(int numBootstraps)	Sets the number of bootstraps to do.
void	setPolyConst(double polyConst)	Sets the constant of the polynomial kernel, if used
void	setPolyDegree(double polyDegree)	Sets the degree of the polynomial kernel, if used
void	setRows(List<Integer> rows)	Allows the user to set which rows are used in the test.
void	setScalingFactor(double scalingFactor)	Sets the width multiplier.
void	setThreshold(double threshold)	Sets the threshold.
void	setVerbose(boolean verbose)	Sets the verbosity of the method.
static org.ejml.simple.SimpleMatrix	standardizeData(org.ejml.simple.SimpleMatrix data)
String	toString()	Returns a string representation of this test.

Methods inherited from class java.lang.Object

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

Methods inherited from interface edu.cmu.tetrad.search.IndependenceTest

checkIndependence, determines, getVariableNames

Constructor Details

Kci

public Kci(DataSet data,
 double alpha)

Constructor.

Parameters:

data - The dataset to analyze. Must be continuous.

alpha - The alpha value of the test.

Method Details

convertTo1DArray

public static double[] convertTo1DArray(org.ejml.simple.SimpleMatrix matrix)

Converts a SimpleMatrix to a 1D array.

Parameters:

matrix - The matrix to convert.

Returns:

The 1D array.

standardizeData

public static org.ejml.simple.SimpleMatrix standardizeData(org.ejml.simple.SimpleMatrix data)

indTestSubset

public IndependenceTest indTestSubset(List<Node> vars)

Description copied from interface: IndependenceTest

Returns an Independence test for a sublist of the variables.

Specified by:

indTestSubset in interface IndependenceTest

Parameters:

vars - The sublist of variables.

Returns:

a IndependenceTest object

Throws:

UnsupportedOperationException - since not implemented.

checkIndependence

public IndependenceResult checkIndependence(Node x,
 Node y,
 Set<Node> z)
                                     throws InterruptedException

Checks the independence between two nodes given a set of conditioning variables.

Specified by:

checkIndependence in interface IndependenceTest

Parameters:

x - The first node.

y - The second node.

z - The set of conditioning variables.

Returns:

The result of the independence test.

Throws:

InterruptedException

See Also:

• IndependenceResult

getVariables

public List<Node> getVariables()

Returns the list of variables over which this independence checker is capable of determinining independence relations.

Specified by:

getVariables in interface IndependenceTest

Returns:

This list.

getVariable

public Node getVariable(String name)

Returns the variable of the given name.

Specified by:

getVariable in interface IndependenceTest

Parameters:

name - a String object representing the name of the variable to retrieve

Returns:

the Node object representing the variable with the given name

getAlpha

public double getAlpha()

Returns the significance level of the independence test.

Specified by:

getAlpha in interface IndependenceTest

Returns:

This alpha.

setAlpha

public void setAlpha(double alpha)

Sets the alpha level for the test.

Specified by:

setAlpha in interface IndependenceTest

Parameters:

alpha - The alpha level to be set.

toString

public String toString()

Returns a string representation of this test.

Specified by:

toString in interface IndependenceTest

Overrides:

toString in class Object

Returns:

This string.

getData

public DataModel getData()

Returns The data model for the independence test.

Specified by:

getData in interface IndependenceTest

Returns:

This data.

See Also:

• DataSet
• ICovarianceMatrix
• DataModel

getCov

public ICovarianceMatrix getCov()

Returns the covariance matrix.

Specified by:

getCov in interface IndependenceTest

Returns:

The covariance matrix.

getDataSets

public List<DataSet> getDataSets()

Returns a list consisting of the dataset for this test.

Specified by:

getDataSets in interface IndependenceTest

Returns:

This dataset in a list.

getSampleSize

public int getSampleSize()

Returns the sample size.

Specified by:

getSampleSize in interface IndependenceTest

Returns:

This size.

getScore

public double getScore(IndependenceResult result)

Returns alpha - p.

Parameters:

result - a IndependenceResult object

Returns:

This number.

setApproximate

public void setApproximate(boolean approximate)

Sets whether the approximate algorithm should be used.

Parameters:

approximate - True, if so.

setScalingFactor

public void setScalingFactor(double scalingFactor)

Sets the width multiplier.

Parameters:

scalingFactor - This multipler.

setNumBootstraps

public void setNumBootstraps(int numBootstraps)

Sets the number of bootstraps to do.

Parameters:

numBootstraps - This number.

setThreshold

public void setThreshold(double threshold)

Sets the threshold.

Parameters:

threshold - This number.

isVerbose

public boolean isVerbose()

Returns the value of the verbose flag.

Specified by:

isVerbose in interface IndependenceTest

Returns:

The value of the verbose flag.

setVerbose

public void setVerbose(boolean verbose)

Sets the verbosity of the method.

Specified by:

setVerbose in interface IndependenceTest

Parameters:

verbose - True if verbosity is enabled, false otherwise.

setKernelType

public void setKernelType(Kci.KernelType kernelType)

Sets the type of kernel to be used in computations.

Parameters:

kernelType - the KernelType to set

setPolyDegree

public void setPolyDegree(double polyDegree)

Sets the degree of the polynomial kernel, if used

Parameters:

polyDegree - the degree of the polynomial kernel to set

setPolyConst

public void setPolyConst(double polyConst)

Sets the constant of the polynomial kernel, if used

Parameters:

polyConst - the constant of the polynomial kernel to set

getRows

public List<Integer> getRows()

Returns the rows used in the test.

Specified by:

getRows in interface RowsSettable

Returns:

The rows used in the test.

setRows

public void setRows(List<Integer> rows)

Allows the user to set which rows are used in the test. Otherwise, all rows are used, except those with missing values.

Specified by:

setRows in interface RowsSettable

Parameters:

rows - The rows to use for the test. Can be null.