Hey Phil,
I notice your moving much of the extended functionality out of the Univariates and into a private field instead. I'm just curious what your logic is behind it? I had considered wrapping vs extending initially when I was building the implementation. Now that I look back, my reasons for going with extending were more to take advantage of the already preexisting "protected" fields in the implementation of the underlying Univariate. Now that I see your changes, I think what your wrapping is more kosher in terms of encapsulation underlying implementation of the wrapped statistic.
While the difference between these approaches is minimal in the front end Univariates like Mean, Var, Median, StandardDev and Geomean, I believe the extended hierarchy of FirstMoment -> SecondMoment -> ThirdMoment -> FourthMoment takes more advantage of this extended field reuse. I'm not so sure if these should be approached as wrappers too? They do make significant reuse of the internal state of the underlying moment. What do you think? Would these be problematic to turn into wrapped implementations too?
-Mark
[EMAIL PROTECTED] wrote:
psteitz 2004/07/17 21:37:08
Modified: math/src/java/org/apache/commons/math/stat/univariate/moment
GeometricMean.java
Log:
Changed implementation to wrap, rather than extend SumOfLogs.
Revision Changes Path
1.22 +30 -20 jakarta-commons/math/src/java/org/apache/commons/math/stat/univariate/moment/GeometricMean.java
Index: GeometricMean.java
===================================================================
RCS file: /home/cvs/jakarta-commons/math/src/java/org/apache/commons/math/stat/univariate/moment/GeometricMean.java,v
retrieving revision 1.21
retrieving revision 1.22
diff -u -r1.21 -r1.22
--- GeometricMean.java 4 Jul 2004 09:02:36 -0000 1.21
+++ GeometricMean.java 18 Jul 2004 04:37:08 -0000 1.22
@@ -15,20 +15,20 @@
*/
package org.apache.commons.math.stat.univariate.moment;
-import java.io.Serializable;
-
+import org.apache.commons.math.stat.univariate.AbstractStorelessUnivariateStatistic;
import org.apache.commons.math.stat.univariate.summary.SumOfLogs;
/**
* Returns the <a href="http://www.xycoon.com/geometric_mean.htm";>
* geometric mean </a> of the available values.
* <p>
- * Uses [EMAIL PROTECTED] SumOfLogs} superclass to compute sum of logs and returns
+ * Uses a [EMAIL PROTECTED] SumOfLogs} instance to compute sum of logs and returns
* <code> exp( 1/n (sum of logs) ).</code> Therefore,
* <ul>
* <li>If any of values are < 0, the result is <code>NaN.</code></li>
- * <li>If all values are non-negative and less than <code>Double.POSITIVE_INFINITY</code>, - * but at least one value is 0, the result is <code>0.</code></li>
+ * <li>If all values are non-negative and less than + * <code>Double.POSITIVE_INFINITY</code>, but at least one value is 0, the + * result is <code>0.</code></li>
* <li>If both <code>Double.POSITIVE_INFINITY</code> and * <code>Double.NEGATIVE_INFINITY</code> are among the values, the result is
* <code>NaN.</code></li>
@@ -42,28 +42,34 @@
*
* @version $Revision$ $Date$
*/
-public class GeometricMean extends SumOfLogs implements Serializable{
+public class GeometricMean extends AbstractStorelessUnivariateStatistic {
/** Serializable version identifier */
static final long serialVersionUID = -8178734905303459453L; - - /**Number of values that have been added */
- protected long n = 0;
+ + /** Wrapped SumOfLogs instance */
+ private SumOfLogs sumOfLogs;
/**
+ * Create a GeometricMean instance
+ */
+ public GeometricMean() {
+ sumOfLogs = new SumOfLogs();
+ }
+ + /**
* @see org.apache.commons.math.stat.univariate.StorelessUnivariateStatistic#increment(double)
*/
public void increment(final double d) {
- n++;
- super.increment(d);
+ sumOfLogs.increment(d);
}
/**
* @see org.apache.commons.math.stat.univariate.StorelessUnivariateStatistic#getResult()
*/
public double getResult() {
- if (n > 0) {
- return Math.exp(super.getResult() / (double) n);
+ if (sumOfLogs.getN() > 0) {
+ return Math.exp(sumOfLogs.getResult() / (double) sumOfLogs.getN());
} else {
return Double.NaN;
}
@@ -73,8 +79,7 @@
* @see org.apache.commons.math.stat.univariate.StorelessUnivariateStatistic#clear()
*/
public void clear() {
- super.clear();
- n = 0;
+ sumOfLogs.clear();
}
/**
@@ -94,11 +99,16 @@
* index parameters are not valid
*/
public double evaluate(
- final double[] values,
- final int begin,
- final int length) {
+ final double[] values, final int begin, final int length) {
return Math.exp(
- super.evaluate(values, begin, length) / (double) length);
+ sumOfLogs.evaluate(values, begin, length) / (double) length);
+ }
+ + /**
+ * @see org.apache.commons.math.stat.univariate.StorelessUnivariateStatistic#getN()
+ */
+ public long getN() {
+ return sumOfLogs.getN();
}
}
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
-- Mark Diggory Software Developer Harvard MIT Data Center http://www.hmdc.harvard.edu
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
