pal.statistics
Class KishinoHasegawaTest

java.lang.Object
  |
  +--pal.statistics.KishinoHasegawaTest
All Implemented Interfaces:
Report

public class KishinoHasegawaTest
extends java.lang.Object
implements Report

Kishino-Hasegawa-(Templeton)-Test (1989, 1983) to compare a set of evolutionary hypotheses


Field Summary
 int bestH
          number of maximum likelihood hypothesis
 double[] delta
          log-likelihood difference to maximum likelihood hypothesis
 double[] deltaSE
          estimated error of log-likelihood differences
 double[] pval
          corresponding p-value (two-sided test on normal distribution)
 
Constructor Summary
KishinoHasegawaTest()
           
 
Method Summary
 void compare(double[][] sLogL)
          Compare all given hypotheses to the best (ML) hypothesis and store results in public arrays delta, deltaSE, pval (which will automatically be created by this procedure).
 void compare(double[][] pLogL, int[] alias)
          Compare all given hypotheses to the best (ML) hypothesis and store results in public arrays delta, deltaSE, pval (which will automatically be created by this procedure).
 void report(java.io.PrintWriter out)
          print human readable report (e.g., on parameters and associated model)
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Field Detail

bestH

public int bestH
number of maximum likelihood hypothesis

delta

public double[] delta
log-likelihood difference to maximum likelihood hypothesis

deltaSE

public double[] deltaSE
estimated error of log-likelihood differences

pval

public double[] pval
corresponding p-value (two-sided test on normal distribution)
Constructor Detail

KishinoHasegawaTest

public KishinoHasegawaTest()
Method Detail

compare

public void compare(double[][] sLogL)
Compare all given hypotheses to the best (ML) hypothesis and store results in public arrays delta, deltaSE, pval (which will automatically be created by this procedure).
Parameters:
sLogL - log-likelihoods of each site

compare

public void compare(double[][] pLogL,
                    int[] alias)
Compare all given hypotheses to the best (ML) hypothesis and store results in public arrays delta, deltaSE, pval (which will automatically be created by this procedure).
Parameters:
pLogL - log-likelihoods of each pattern
alias - map of patterns to sites in sequence

report

public void report(java.io.PrintWriter out)
Description copied from interface: Report
print human readable report (e.g., on parameters and associated model)
Specified by:
report in interface Report
Following copied from interface: pal.misc.Report
Parameters:
out - output stream