Package edu.cmu.tetrad.search

Class Cdnod

java.lang.Object

edu.cmu.tetrad.search.Cdnod

All Implemented Interfaces:

IGraphSearch

public final class Cdnod
extends Object
implements IGraphSearch

The Cdnod class implements the causal discovery algorithm for detecting changing dependencies with respect to a change index variable in a dataset.

The algorithm is capable of orienting edges in a graph based on statistical tests, knowledge constraints, and user-defined parameters. It also supports conservative decision-making and constraints on collider orientations.

This class extends the functionality of IGraphSearch and provides methods for configuring the statistical independence test, working dataset, maximum conditioning depth, and timeout settings.

Key Features:

• Searches for a graph representing causal structure based on a change index variable.
• Provides support for collider orientation with both conservative logic and maximum p-margin decisions.
• Enables customization of constraints using a Knowledge object.
• Supports timeout settings to limit the duration of computations.

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static final class	Cdnod.Builder	Builder class for creating instances of the Cdnod class with customized parameters.
static enum	Cdnod.ColliderOrientationStyle	Enumeration representing different strategies for orienting colliders in causal discovery.

Method Summary

Modifier and Type	Method	Description
IndependenceTest	getTest()	Gets the test used by the search.
Graph	search()	Runs the search and returns a graph.
void	setDataAndIndex(DataSet dataX, double[] cIndex, String cName)	Updates the internal dataset by appending a change index as the last column.
void	setDataWithC(DataSet dataWithC)	Sets the dataset to be used in this instance.
void	setTest(IndependenceTest newTest)	Sets the test to be used by the search.
void	setTimeoutMs(long timeoutMs)	Sets the timeout value in milliseconds for this instance.

Methods inherited from class java.lang.Object

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

Method Details

search

public Graph search()
             throws InterruptedException

Description copied from interface: IGraphSearch

Runs the search and returns a graph.

Specified by:

search in interface IGraphSearch

Returns:

The discovered graph.

Throws:

InterruptedException - if any.

getTest

public IndependenceTest getTest()

Description copied from interface: IGraphSearch

Gets the test used by the search.

Specified by:

getTest in interface IGraphSearch

Returns:

The test used by the search.

setTest

public void setTest(IndependenceTest newTest)

Description copied from interface: IGraphSearch

Sets the test to be used by the search. The list of variables of the new proposed test must be equal to the list of variables of the existing test.

Specified by:

setTest in interface IGraphSearch

Parameters:

newTest - The test to be used by the search.

setDataWithC

public void setDataWithC(DataSet dataWithC)

Sets the dataset to be used in this instance.

Parameters:

dataWithC - the dataset to be assigned

setDataAndIndex

public void setDataAndIndex(DataSet dataX,
 double[] cIndex,
 String cName)

Updates the internal dataset by appending a change index as the last column. This method modifies the data to include an additional column (defined by the provided change index and name) and stores the updated dataset for further processing.

Parameters:

dataX - the original dataset to which the change index will be appended

cIndex - an array representing the change index values to be incorporated into the dataset

cName - the name of the new column that will represent the change index

setTimeoutMs

public void setTimeoutMs(long timeoutMs)

Sets the timeout value in milliseconds for this instance. This value determines the maximum time allowed for certain operations before they are interrupted or terminated.

Parameters:

timeoutMs - the timeout in milliseconds