changeset 2645:4f7d1ea38404

Added simple Grubb's outlier test. flys-artifacts/trunk@4300 c6561f87-3c4e-4783-a992-168aeb5c3f6f
author Sascha L. Teichmann <sascha.teichmann@intevation.de>
date Wed, 25 Apr 2012 15:57:23 +0000
parents 0a84313efe60
children c11da3540b70
files flys-artifacts/ChangeLog flys-artifacts/src/main/java/de/intevation/flys/artifacts/math/Outlier.java
diffstat 2 files changed, 100 insertions(+), 1 deletions(-) [+]
line wrap: on
line diff
--- a/flys-artifacts/ChangeLog	Wed Apr 25 15:15:39 2012 +0000
+++ b/flys-artifacts/ChangeLog	Wed Apr 25 15:57:23 2012 +0000
@@ -1,8 +1,13 @@
+2012-04-25	Sascha L. Teichmann	<sascha.teichmann@intevation.de>
+
+	* src/main/java/de/intevation/flys/artifacts/math/Outlier.java:
+	  New. Simple Grubb's outlier test. Needs testing. ;-)
+
 2012-04-25	Felix Wolfsteller	<felix.wolfsteller@intevation.de>
 
 	* src/main/java/de/intevation/flys/jfree/EnhancedLineAndShapeRenderer.java
 	  (isMinimumX): Implemented minimum finding for x in analogy to the ones for
-          y values. Stub to label dataset curves in plot.
+	  y values. Stub to label dataset curves in plot.
 
 2012-04-25	Felix Wolfsteller	<felix.wolfsteller@intevation.de>
 
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/flys-artifacts/src/main/java/de/intevation/flys/artifacts/math/Outlier.java	Wed Apr 25 15:57:23 2012 +0000
@@ -0,0 +1,94 @@
+package de.intevation.flys.artifacts.math;
+
+import org.apache.commons.math.stat.descriptive.moment.Mean;
+import org.apache.commons.math.stat.descriptive.moment.StandardDeviation;
+
+import org.apache.commons.math.distribution.TDistributionImpl;
+
+import java.util.List;
+import java.util.ArrayList;
+
+public class Outlier
+{
+    public static class IndexedValue {
+        protected int    index;
+        protected double value;
+
+        public IndexedValue() {
+        }
+
+        public IndexedValue(int index, double value) {
+            this.index = index;
+            this.value = value;
+        }
+
+        public int getIndex() {
+            return index;
+        }
+
+        public void setIndex(int index) {
+            this.index = index;
+        }
+
+        public double getValue() {
+            return value;
+        }
+
+        public void setValue(double value) {
+            this.value = value;
+        }
+    } // class IndexedValue
+
+    public Outlier() {
+    }
+
+    public static List<IndexedValue> findOutliers(
+        List<IndexedValue> inputValues,
+        double             alpha
+    ) {
+        ArrayList<IndexedValue> outliers = new ArrayList<IndexedValue>();
+
+        ArrayList<IndexedValue> values =
+            new ArrayList<IndexedValue>(inputValues);
+
+        for (;;) {
+            int N = values.size();
+
+            if (N < 4) {
+                break;
+            }
+
+            Mean mean = new Mean();
+            StandardDeviation std = new StandardDeviation();
+
+            for (IndexedValue value: values) {
+                mean.increment(value.getValue());
+                std.increment(value.getValue());
+            }
+
+            double m = mean.getResult();
+            double s = std.getResult();
+
+            double maxZ = -Double.MAX_VALUE;
+            int iv = -1;
+            for (int i >= 0; i = N-1; --i) {
+                IndexedValue v = values.get(i);
+                double z = Math.abs(m - v.getValue())/s;
+                if (z > maxZ) {
+                    maxZ = z;
+                    iv = i;
+                }
+            }
+
+            double t = Math.sqrt((N*(N-2)*z*z)/((N-1)*(N-1) - N*z*z))
+
+            TDistributionImpl tdist = new TDistributionImpl(N-2);
+
+            double p = tdist.cumulativeProbability(t)
+        }
+
+
+        return outliers;
+    }
+}
+// vim:set ts=4 sw=4 si et sta sts=4 fenc=utf8 :

http://dive4elements.wald.intevation.org