SuanShu, a Java numerical and statistical library

com.numericalmethod.suanshu.stats.cointegration
Class JohansenTest

java.lang.Object
  extended by com.numericalmethod.suanshu.stats.cointegration.JohansenTest

public class JohansenTest
extends java.lang.Object

This class provides the Johansen distributions of specific types up to a certain dimension. It is used to perform multiple similar Johansen tests without having to reconstruct the distributions repeatedly.

This class also provides an easy to use wrapper around the CointegrationMle object to extract useful information such as:


Constructor Summary
JohansenTest(CointegrationMle cointegrationMle)
          Construct a specific JohansenTest appropriate for a particular CointegrationMle object.
JohansenTest(JohansenAsymptoticDistribution.Test test, JohansenAsymptoticDistribution.TrendType trend, int dim)
          Construct a specific JohansenTest.
JohansenTest(JohansenAsymptoticDistribution.Test test, JohansenAsymptoticDistribution.TrendType trend, int dim, int nSim, int nT)
          Construct a specific JohansenTest.
 
Method Summary
 int r(CointegrationMle cointegrationMle, double level)
          Get the (most likely) order of cointegration.
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

JohansenTest

public JohansenTest(JohansenAsymptoticDistribution.Test test,
                    JohansenAsymptoticDistribution.TrendType trend,
                    int dim,
                    int nSim,
                    int nT)
Construct a specific JohansenTest.

Parameters:
test - the types of Johansen's cointegration test
trend - the types of trends
dim - the number of (real) eigenvalues
nSim - the number of simulations
nT - the number of discretizations

JohansenTest

public JohansenTest(JohansenAsymptoticDistribution.Test test,
                    JohansenAsymptoticDistribution.TrendType trend,
                    int dim)
Construct a specific JohansenTest.

Parameters:
test - the types of Johansen's cointegration test
trend - the types of trends
dim - the number of (real) eigenvalues

JohansenTest

public JohansenTest(CointegrationMle cointegrationMle)
Construct a specific JohansenTest appropriate for a particular CointegrationMle object.

Parameters:
cointegrationMle - a CointegrationMle object
Method Detail

r

public int r(CointegrationMle cointegrationMle,
             double level)
Get the (most likely) order of cointegration.

Parameters:
cointegrationMle - the CointegrationMle object
level - the confidence level, i.e., 1 - quantile
Returns:
the order of cointegration

SuanShu, a Java numerical and statistical library

Copyright © 2011 Numerical Method Inc. Ltd. All Rights Reserved.