1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16 package org.apache.commons.math.stat.descriptive;
17
18 import java.io.Serializable;
19 import java.util.ArrayList;
20 import java.util.List;
21
22 import org.apache.commons.math.MathException;
23 import org.apache.commons.math.stat.descriptive.UnivariateStatistic;
24 import org.apache.commons.math.stat.descriptive.DescriptiveStatistics;
25 import org.apache.commons.math.util.DefaultTransformer;
26 import org.apache.commons.math.util.NumberTransformer;
27
28
29
30
31 public class ListUnivariateImpl extends DescriptiveStatistics implements Serializable {
32
33
34 private static final long serialVersionUID = -8837442489133392138L;
35
36
37
38
39
40 protected List list;
41
42
43 protected NumberTransformer transformer;
44
45
46 protected int windowSize = DescriptiveStatistics.INFINITE_WINDOW;
47
48
49
50
51 public ListUnivariateImpl(){
52 this(new ArrayList());
53 }
54
55
56
57
58
59 public ListUnivariateImpl(List list) {
60 this(list, new DefaultTransformer());
61 }
62
63
64
65
66
67
68 public ListUnivariateImpl(List list, NumberTransformer transformer) {
69 super();
70 this.list = list;
71 this.transformer = transformer;
72 }
73
74
75
76
77 public double[] getValues() {
78
79 int length = list.size();
80
81
82
83
84
85
86 if (windowSize != DescriptiveStatistics.INFINITE_WINDOW &&
87 windowSize < list.size())
88 {
89 length = list.size() - Math.max(0, list.size() - windowSize);
90 }
91
92
93 double[] copiedArray = new double[length];
94
95 for (int i = 0; i < copiedArray.length; i++) {
96 copiedArray[i] = getElement(i);
97 }
98 return copiedArray;
99 }
100
101
102
103
104 public double getElement(int index) {
105
106 double value = Double.NaN;
107
108 int calcIndex = index;
109
110 if (windowSize != DescriptiveStatistics.INFINITE_WINDOW &&
111 windowSize < list.size())
112 {
113 calcIndex = (list.size() - windowSize) + index;
114 }
115
116
117 try {
118 value = transformer.transform(list.get(calcIndex));
119 } catch (MathException e) {
120 e.printStackTrace();
121 }
122
123 return value;
124 }
125
126
127
128
129 public long getN() {
130 int n = 0;
131
132 if (windowSize != DescriptiveStatistics.INFINITE_WINDOW) {
133 if (list.size() > windowSize) {
134 n = windowSize;
135 } else {
136 n = list.size();
137 }
138 } else {
139 n = list.size();
140 }
141 return n;
142 }
143
144
145
146
147 public void addValue(double v) {
148 list.add(new Double(v));
149 }
150
151
152
153
154
155 public void addObject(Object o) {
156 list.add(o);
157 }
158
159
160
161
162
163
164 public void clear() {
165 list.clear();
166 }
167
168
169
170
171
172
173 public double apply(UnivariateStatistic stat) {
174 double[] v = this.getValues();
175
176 if (v != null) {
177 return stat.evaluate(v, 0, v.length);
178 }
179 return Double.NaN;
180 }
181
182
183
184
185
186 public NumberTransformer getTransformer() {
187 return transformer;
188 }
189
190
191
192
193
194 public void setTransformer(NumberTransformer transformer) {
195 this.transformer = transformer;
196 }
197
198
199
200
201 public synchronized void setWindowSize(int windowSize) {
202 this.windowSize = windowSize;
203
204
205 int extra = list.size() - windowSize;
206 for (int i = 0; i < extra; i++) {
207 list.remove(0);
208 }
209 }
210
211
212
213
214 public int getWindowSize() {
215 return windowSize;
216 }
217
218 }