Hello,
The org.apache.commons.math.stat.* packages are exactly what we need for our
microarray analysis web application except for the lack of an ANOVA class (the
FDistribution is already there however). I have coded this up and it seems to
produce the same output as R's anova(lm(data)) - as described in more detail
in the test suite.
I have added the following to my anonymous SVN checkout:
src/test/org/apache/commons/math/stat/inference/AnovaTest.java
src/java/org/apache/commons/math/stat/inference/Anova.java
src/java/org/apache/commons/math/stat/inference/AnovaImpl.java
Should I add this to a bugzilla ticket for you guys to look at? How do I get
a suitable patch file? If I do svn diff, these new files are not in
subversion... Or should I just tar up these files?
I guess one issue is naming... Should it be called UnivariateAnova, for
example?
Here are the methods from interface definition as a taster.
As you can see, I need to flesh out the documentation some more.
/**
* Computes the ANOVA F-value for a set of data in two or more classes.
*
* @param classdata Hashtable - key = classname, values = double[]
* @return Fvalue
* @throws IllegalArgumentException if the precondition is not met
* @throws MathException if the statistic can not be computed do to a
* convergence or other numerical error.
*/
public abstract double anovaFValue(Hashtable classdata)
throws IllegalArgumentException, MathException;
/**
* Computes the ANOVA P-value for a set of data in two or more classes.
*
* @param classdata Hashtable - key = classname, values = double[]
* @return Pvalue
* @throws IllegalArgumentException if the precondition is not met
* @throws MathException if the statistic can not be computed do to a
* convergence or other numerical error.
*/
public abstract double anovaPValue(Hashtable classdata)
throws IllegalArgumentException, MathException;
I have followed TTest as a style guide.
cheers,
Bob.
--
Bob MacCallum | VectorBase Developer | Kafatos/Christophides Groups |
Division of Cell and Molecular Biology | Imperial College London |
Phone +442075941945 | Email [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
