1 | #region License Information
|
---|
2 | /* HeuristicLab
|
---|
3 | * Copyright (C) 2002-2018 Heuristic and Evolutionary Algorithms Laboratory (HEAL)
|
---|
4 | *
|
---|
5 | * This file is part of HeuristicLab.
|
---|
6 | *
|
---|
7 | * HeuristicLab is free software: you can redistribute it and/or modify
|
---|
8 | * it under the terms of the GNU General Public License as published by
|
---|
9 | * the Free Software Foundation, either version 3 of the License, or
|
---|
10 | * (at your option) any later version.
|
---|
11 | *
|
---|
12 | * HeuristicLab is distributed in the hope that it will be useful,
|
---|
13 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
14 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
15 | * GNU General Public License for more details.
|
---|
16 | *
|
---|
17 | * You should have received a copy of the GNU General Public License
|
---|
18 | * along with HeuristicLab. If not, see <http://www.gnu.org/licenses/>.
|
---|
19 | */
|
---|
20 | #endregion
|
---|
21 |
|
---|
22 | using System;
|
---|
23 | using System.Collections.Generic;
|
---|
24 | using System.Linq;
|
---|
25 | using HeuristicLab.Common;
|
---|
26 | using HeuristicLab.Core;
|
---|
27 | using HeuristicLab.Data;
|
---|
28 | using HeuristicLab.Optimization;
|
---|
29 | using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
|
---|
30 |
|
---|
31 | namespace HeuristicLab.Problems.DataAnalysis {
|
---|
32 | [StorableClass]
|
---|
33 | [Item("Prognosis Results", "Represents a collection of time series prognosis results.")]
|
---|
34 | public class TimeSeriesPrognosisResults : ResultCollection {
|
---|
35 | #region result names
|
---|
36 | protected const string PrognosisTrainingMeanSquaredErrorResultName = "Mean squared error (training)";
|
---|
37 | protected const string PrognosisTestMeanSquaredErrorResultName = "Mean squared error (test)";
|
---|
38 | protected const string PrognosisTrainingMeanAbsoluteErrorResultName = "Mean absolute error (training)";
|
---|
39 | protected const string PrognosisTestMeanAbsoluteErrorResultName = "Mean absolute error (test)";
|
---|
40 | protected const string PrognosisTrainingSquaredCorrelationResultName = "Pearson's R² (training)";
|
---|
41 | protected const string PrognosisTestSquaredCorrelationResultName = "Pearson's R² (test)";
|
---|
42 | protected const string PrognosisTrainingRelativeErrorResultName = "Average relative error (training)";
|
---|
43 | protected const string PrognosisTestRelativeErrorResultName = "Average relative error (test)";
|
---|
44 | protected const string PrognosisTrainingNormalizedMeanSquaredErrorResultName = "Normalized mean squared error (training)";
|
---|
45 | protected const string PrognosisTestNormalizedMeanSquaredErrorResultName = "Normalized mean squared error (test)";
|
---|
46 | protected const string PrognosisTrainingMeanErrorResultName = "Mean error (training)";
|
---|
47 | protected const string PrognosisTestMeanErrorResultName = "Mean error (test)";
|
---|
48 |
|
---|
49 | protected const string PrognosisTrainingDirectionalSymmetryResultName = "Average directional symmetry (training)";
|
---|
50 | protected const string PrognosisTestDirectionalSymmetryResultName = "Average directional symmetry (test)";
|
---|
51 | protected const string PrognosisTrainingWeightedDirectionalSymmetryResultName = "Average weighted directional symmetry (training)";
|
---|
52 | protected const string PrognosisTestWeightedDirectionalSymmetryResultName = "Average weighted directional symmetry (test)";
|
---|
53 | protected const string PrognosisTrainingTheilsUStatisticAR1ResultName = "Theil's U2 (AR1) (training)";
|
---|
54 | protected const string PrognosisTestTheilsUStatisticAR1ResultName = "Theil's U2 (AR1) (test)";
|
---|
55 | protected const string PrognosisTrainingTheilsUStatisticMeanResultName = "Theil's U2 (mean) (training)";
|
---|
56 | protected const string PrognosisTestTheilsUStatisticMeanResultName = "Theil's U2 (mean) (test)";
|
---|
57 | #endregion
|
---|
58 |
|
---|
59 | #region result descriptions
|
---|
60 | protected const string PrognosisTrainingMeanSquaredErrorResultDescription = "Mean of squared errors of the model on the training partition";
|
---|
61 | protected const string PrognosisTestMeanSquaredErrorResultDescription = "Mean of squared errors of the model on the test partition";
|
---|
62 | protected const string PrognosisTrainingMeanAbsoluteErrorResultDescription = "Mean of absolute errors of the model on the training partition";
|
---|
63 | protected const string PrognosisTestMeanAbsoluteErrorResultDescription = "Mean of absolute errors of the model on the test partition";
|
---|
64 | protected const string PrognosisTrainingSquaredCorrelationResultDescription = "Squared Pearson's correlation coefficient of the model output and the actual values on the training partition";
|
---|
65 | protected const string PrognosisTestSquaredCorrelationResultDescription = "Squared Pearson's correlation coefficient of the model output and the actual values on the test partition";
|
---|
66 | protected const string PrognosisTrainingRelativeErrorResultDescription = "Average of the relative errors of the model output and the actual values on the training partition";
|
---|
67 | protected const string PrognosisTestRelativeErrorResultDescription = "Average of the relative errors of the model output and the actual values on the test partition";
|
---|
68 | protected const string PrognosisTrainingNormalizedMeanSquaredErrorResultDescription = "Normalized mean of squared errors of the model on the training partition";
|
---|
69 | protected const string PrognosisTestNormalizedMeanSquaredErrorResultDescription = "Normalized mean of squared errors of the model on the test partition";
|
---|
70 | protected const string PrognosisTrainingMeanErrorResultDescription = "Mean of errors of the model on the training partition";
|
---|
71 | protected const string PrognosisTestMeanErrorResultDescription = "Mean of errors of the model on the test partition";
|
---|
72 |
|
---|
73 | protected const string PrognosisTrainingDirectionalSymmetryResultDescription = "The average directional symmetry of the forecasts of the model on the training partition";
|
---|
74 | protected const string PrognosisTestDirectionalSymmetryResultDescription = "The average directional symmetry of the forecasts of the model on the test partition";
|
---|
75 | protected const string PrognosisTrainingWeightedDirectionalSymmetryResultDescription = "The average weighted directional symmetry of the forecasts of the model on the training partition";
|
---|
76 | protected const string PrognosisTestWeightedDirectionalSymmetryResultDescription = "The average weighted directional symmetry of the forecasts of the model on the test partition";
|
---|
77 | protected const string PrognosisTrainingTheilsUStatisticAR1ResultDescription = "The Theil's U statistic (reference: AR1 model) of the forecasts of the model on the training partition";
|
---|
78 | protected const string PrognosisTestTheilsUStatisticAR1ResultDescription = "The Theil's U statistic (reference: AR1 model) of the forecasts of the model on the test partition";
|
---|
79 | protected const string PrognosisTrainingTheilsUStatisticMeanResultDescription = "The Theil's U statistic (reference: mean model) of the forecasts of the model on the training partition";
|
---|
80 | protected const string PrognosisTestTheilsUStatisticMeanResultDescription = "The Theil's U statistic (reference: mean value) of the forecasts of the model on the test partition";
|
---|
81 | #endregion
|
---|
82 |
|
---|
83 | #region result properties
|
---|
84 | //prognosis results for different horizons
|
---|
85 | public double PrognosisTrainingMeanSquaredError {
|
---|
86 | get {
|
---|
87 | if (!ContainsKey(PrognosisTrainingMeanSquaredErrorResultName)) return double.NaN;
|
---|
88 | return ((DoubleValue)this[PrognosisTrainingMeanSquaredErrorResultName].Value).Value;
|
---|
89 | }
|
---|
90 | private set {
|
---|
91 | if (!ContainsKey(PrognosisTrainingMeanSquaredErrorResultName)) Add(new Result(PrognosisTrainingMeanSquaredErrorResultName, PrognosisTrainingMeanSquaredErrorResultDescription, new DoubleValue()));
|
---|
92 | ((DoubleValue)this[PrognosisTrainingMeanSquaredErrorResultName].Value).Value = value;
|
---|
93 | }
|
---|
94 | }
|
---|
95 |
|
---|
96 | public double PrognosisTestMeanSquaredError {
|
---|
97 | get {
|
---|
98 | if (!ContainsKey(PrognosisTestMeanSquaredErrorResultName)) return double.NaN;
|
---|
99 | return ((DoubleValue)this[PrognosisTestMeanSquaredErrorResultName].Value).Value;
|
---|
100 | }
|
---|
101 | private set {
|
---|
102 | if (!ContainsKey(PrognosisTestMeanSquaredErrorResultName)) Add(new Result(PrognosisTestMeanSquaredErrorResultName, PrognosisTestMeanSquaredErrorResultDescription, new DoubleValue()));
|
---|
103 | ((DoubleValue)this[PrognosisTestMeanSquaredErrorResultName].Value).Value = value;
|
---|
104 | }
|
---|
105 | }
|
---|
106 |
|
---|
107 | public double PrognosisTrainingMeanAbsoluteError {
|
---|
108 | get {
|
---|
109 | if (!ContainsKey(PrognosisTrainingMeanAbsoluteErrorResultName)) return double.NaN;
|
---|
110 | return ((DoubleValue)this[PrognosisTrainingMeanAbsoluteErrorResultName].Value).Value;
|
---|
111 | }
|
---|
112 | private set {
|
---|
113 | if (!ContainsKey(PrognosisTrainingMeanAbsoluteErrorResultName)) Add(new Result(PrognosisTrainingMeanAbsoluteErrorResultName, PrognosisTrainingMeanAbsoluteErrorResultDescription, new DoubleValue()));
|
---|
114 | ((DoubleValue)this[PrognosisTrainingMeanAbsoluteErrorResultName].Value).Value = value;
|
---|
115 | }
|
---|
116 | }
|
---|
117 |
|
---|
118 | public double PrognosisTestMeanAbsoluteError {
|
---|
119 | get {
|
---|
120 | if (!ContainsKey(PrognosisTestMeanAbsoluteErrorResultName)) return double.NaN;
|
---|
121 | return ((DoubleValue)this[PrognosisTestMeanAbsoluteErrorResultName].Value).Value;
|
---|
122 | }
|
---|
123 | private set {
|
---|
124 | if (!ContainsKey(PrognosisTestMeanAbsoluteErrorResultName)) Add(new Result(PrognosisTestMeanAbsoluteErrorResultName, PrognosisTestMeanAbsoluteErrorResultDescription, new DoubleValue()));
|
---|
125 | ((DoubleValue)this[PrognosisTestMeanAbsoluteErrorResultName].Value).Value = value;
|
---|
126 | }
|
---|
127 | }
|
---|
128 |
|
---|
129 | public double PrognosisTrainingRSquared {
|
---|
130 | get {
|
---|
131 | if (!ContainsKey(PrognosisTrainingSquaredCorrelationResultName)) return double.NaN;
|
---|
132 | return ((DoubleValue)this[PrognosisTrainingSquaredCorrelationResultName].Value).Value;
|
---|
133 | }
|
---|
134 | private set {
|
---|
135 | if (!ContainsKey(PrognosisTrainingSquaredCorrelationResultName)) Add(new Result(PrognosisTrainingSquaredCorrelationResultName, PrognosisTrainingSquaredCorrelationResultDescription, new DoubleValue()));
|
---|
136 | ((DoubleValue)this[PrognosisTrainingSquaredCorrelationResultName].Value).Value = value;
|
---|
137 | }
|
---|
138 | }
|
---|
139 |
|
---|
140 | public double PrognosisTestRSquared {
|
---|
141 | get {
|
---|
142 | if (!ContainsKey(PrognosisTestSquaredCorrelationResultName)) return double.NaN;
|
---|
143 | return ((DoubleValue)this[PrognosisTestSquaredCorrelationResultName].Value).Value;
|
---|
144 | }
|
---|
145 | private set {
|
---|
146 | if (!ContainsKey(PrognosisTestSquaredCorrelationResultName)) Add(new Result(PrognosisTestSquaredCorrelationResultName, PrognosisTestSquaredCorrelationResultDescription, new DoubleValue()));
|
---|
147 | ((DoubleValue)this[PrognosisTestSquaredCorrelationResultName].Value).Value = value;
|
---|
148 | }
|
---|
149 | }
|
---|
150 |
|
---|
151 | public double PrognosisTrainingRelativeError {
|
---|
152 | get {
|
---|
153 | if (!ContainsKey(PrognosisTrainingRelativeErrorResultName)) return double.NaN;
|
---|
154 | return ((DoubleValue)this[PrognosisTrainingRelativeErrorResultName].Value).Value;
|
---|
155 | }
|
---|
156 | private set {
|
---|
157 | if (!ContainsKey(PrognosisTrainingRelativeErrorResultName)) Add(new Result(PrognosisTrainingRelativeErrorResultName, PrognosisTrainingRelativeErrorResultDescription, new DoubleValue()));
|
---|
158 | ((DoubleValue)this[PrognosisTrainingRelativeErrorResultName].Value).Value = value;
|
---|
159 | }
|
---|
160 | }
|
---|
161 |
|
---|
162 | public double PrognosisTestRelativeError {
|
---|
163 | get {
|
---|
164 | if (!ContainsKey(PrognosisTestRelativeErrorResultName)) return double.NaN;
|
---|
165 | return ((DoubleValue)this[PrognosisTestRelativeErrorResultName].Value).Value;
|
---|
166 | }
|
---|
167 | private set {
|
---|
168 | if (!ContainsKey(PrognosisTestRelativeErrorResultName)) Add(new Result(PrognosisTestRelativeErrorResultName, PrognosisTestRelativeErrorResultDescription, new DoubleValue()));
|
---|
169 | ((DoubleValue)this[PrognosisTestRelativeErrorResultName].Value).Value = value;
|
---|
170 | }
|
---|
171 | }
|
---|
172 |
|
---|
173 | public double PrognosisTrainingNormalizedMeanSquaredError {
|
---|
174 | get {
|
---|
175 | if (!ContainsKey(PrognosisTrainingNormalizedMeanSquaredErrorResultName)) return double.NaN;
|
---|
176 | return ((DoubleValue)this[PrognosisTrainingNormalizedMeanSquaredErrorResultName].Value).Value;
|
---|
177 | }
|
---|
178 | private set {
|
---|
179 | if (!ContainsKey(PrognosisTrainingNormalizedMeanSquaredErrorResultName)) Add(new Result(PrognosisTrainingNormalizedMeanSquaredErrorResultName, PrognosisTrainingNormalizedMeanSquaredErrorResultDescription, new DoubleValue()));
|
---|
180 | ((DoubleValue)this[PrognosisTrainingNormalizedMeanSquaredErrorResultName].Value).Value = value;
|
---|
181 | }
|
---|
182 | }
|
---|
183 |
|
---|
184 | public double PrognosisTestNormalizedMeanSquaredError {
|
---|
185 | get {
|
---|
186 | if (!ContainsKey(PrognosisTestNormalizedMeanSquaredErrorResultName)) return double.NaN;
|
---|
187 | return ((DoubleValue)this[PrognosisTestNormalizedMeanSquaredErrorResultName].Value).Value;
|
---|
188 | }
|
---|
189 | private set {
|
---|
190 | if (!ContainsKey(PrognosisTestNormalizedMeanSquaredErrorResultName)) Add(new Result(PrognosisTestNormalizedMeanSquaredErrorResultName, PrognosisTestNormalizedMeanSquaredErrorResultDescription, new DoubleValue()));
|
---|
191 | ((DoubleValue)this[PrognosisTestNormalizedMeanSquaredErrorResultName].Value).Value = value;
|
---|
192 | }
|
---|
193 | }
|
---|
194 |
|
---|
195 | public double PrognosisTrainingMeanError {
|
---|
196 | get {
|
---|
197 | if (!ContainsKey(PrognosisTrainingMeanErrorResultName)) return double.NaN;
|
---|
198 | return ((DoubleValue)this[PrognosisTrainingMeanErrorResultName].Value).Value;
|
---|
199 | }
|
---|
200 | private set {
|
---|
201 | if (!ContainsKey(PrognosisTrainingMeanErrorResultName)) Add(new Result(PrognosisTrainingMeanErrorResultName, PrognosisTrainingMeanErrorResultDescription, new DoubleValue()));
|
---|
202 | ((DoubleValue)this[PrognosisTrainingMeanErrorResultName].Value).Value = value;
|
---|
203 | }
|
---|
204 | }
|
---|
205 |
|
---|
206 | public double PrognosisTestMeanError {
|
---|
207 | get {
|
---|
208 | if (!ContainsKey(PrognosisTestMeanErrorResultName)) return double.NaN;
|
---|
209 | return ((DoubleValue)this[PrognosisTestMeanErrorResultName].Value).Value;
|
---|
210 | }
|
---|
211 | private set {
|
---|
212 | if (!ContainsKey(PrognosisTestMeanErrorResultName)) Add(new Result(PrognosisTestMeanErrorResultName, PrognosisTestMeanErrorResultDescription, new DoubleValue()));
|
---|
213 | ((DoubleValue)this[PrognosisTestMeanErrorResultName].Value).Value = value;
|
---|
214 | }
|
---|
215 | }
|
---|
216 |
|
---|
217 |
|
---|
218 | public double PrognosisTrainingDirectionalSymmetry {
|
---|
219 | get {
|
---|
220 | if (!ContainsKey(PrognosisTrainingDirectionalSymmetryResultName)) return double.NaN;
|
---|
221 | return ((DoubleValue)this[PrognosisTrainingDirectionalSymmetryResultName].Value).Value;
|
---|
222 | }
|
---|
223 | private set {
|
---|
224 | if (!ContainsKey(PrognosisTrainingDirectionalSymmetryResultName)) Add(new Result(PrognosisTrainingDirectionalSymmetryResultName, PrognosisTrainingDirectionalSymmetryResultDescription, new DoubleValue()));
|
---|
225 | ((DoubleValue)this[PrognosisTrainingDirectionalSymmetryResultName].Value).Value = value;
|
---|
226 | }
|
---|
227 | }
|
---|
228 | public double PrognosisTestDirectionalSymmetry {
|
---|
229 | get {
|
---|
230 | if (!ContainsKey(PrognosisTestDirectionalSymmetryResultName)) return double.NaN;
|
---|
231 | return ((DoubleValue)this[PrognosisTestDirectionalSymmetryResultName].Value).Value;
|
---|
232 | }
|
---|
233 | private set {
|
---|
234 | if (!ContainsKey(PrognosisTestDirectionalSymmetryResultName)) Add(new Result(PrognosisTestDirectionalSymmetryResultName, PrognosisTestDirectionalSymmetryResultDescription, new DoubleValue()));
|
---|
235 | ((DoubleValue)this[PrognosisTestDirectionalSymmetryResultName].Value).Value = value;
|
---|
236 | }
|
---|
237 | }
|
---|
238 | public double PrognosisTrainingWeightedDirectionalSymmetry {
|
---|
239 | get {
|
---|
240 | if (!ContainsKey(PrognosisTrainingWeightedDirectionalSymmetryResultName)) return double.NaN;
|
---|
241 | return ((DoubleValue)this[PrognosisTrainingWeightedDirectionalSymmetryResultName].Value).Value;
|
---|
242 | }
|
---|
243 | private set {
|
---|
244 | if (!ContainsKey(PrognosisTrainingWeightedDirectionalSymmetryResultName)) Add(new Result(PrognosisTrainingWeightedDirectionalSymmetryResultName, PrognosisTrainingWeightedDirectionalSymmetryResultDescription, new DoubleValue()));
|
---|
245 | ((DoubleValue)this[PrognosisTrainingWeightedDirectionalSymmetryResultName].Value).Value = value;
|
---|
246 | }
|
---|
247 | }
|
---|
248 | public double PrognosisTestWeightedDirectionalSymmetry {
|
---|
249 | get {
|
---|
250 | if (!ContainsKey(PrognosisTestWeightedDirectionalSymmetryResultName)) return double.NaN;
|
---|
251 | return ((DoubleValue)this[PrognosisTestWeightedDirectionalSymmetryResultName].Value).Value;
|
---|
252 | }
|
---|
253 | private set {
|
---|
254 | if (!ContainsKey(PrognosisTestWeightedDirectionalSymmetryResultName)) Add(new Result(PrognosisTestWeightedDirectionalSymmetryResultName, PrognosisTestWeightedDirectionalSymmetryResultDescription, new DoubleValue()));
|
---|
255 | ((DoubleValue)this[PrognosisTestWeightedDirectionalSymmetryResultName].Value).Value = value;
|
---|
256 | }
|
---|
257 | }
|
---|
258 | public double PrognosisTrainingTheilsUStatisticAR1 {
|
---|
259 | get {
|
---|
260 | if (!ContainsKey(PrognosisTrainingTheilsUStatisticAR1ResultName)) return double.NaN;
|
---|
261 | return ((DoubleValue)this[PrognosisTrainingTheilsUStatisticAR1ResultName].Value).Value;
|
---|
262 | }
|
---|
263 | private set {
|
---|
264 | if (!ContainsKey(PrognosisTrainingTheilsUStatisticAR1ResultName)) Add(new Result(PrognosisTrainingTheilsUStatisticAR1ResultName, PrognosisTrainingTheilsUStatisticAR1ResultDescription, new DoubleValue()));
|
---|
265 | ((DoubleValue)this[PrognosisTrainingTheilsUStatisticAR1ResultName].Value).Value = value;
|
---|
266 | }
|
---|
267 | }
|
---|
268 | public double PrognosisTestTheilsUStatisticAR1 {
|
---|
269 | get {
|
---|
270 | if (!ContainsKey(PrognosisTestTheilsUStatisticAR1ResultName)) return double.NaN;
|
---|
271 | return ((DoubleValue)this[PrognosisTestTheilsUStatisticAR1ResultName].Value).Value;
|
---|
272 | }
|
---|
273 | private set {
|
---|
274 | if (!ContainsKey(PrognosisTestTheilsUStatisticAR1ResultName)) Add(new Result(PrognosisTestTheilsUStatisticAR1ResultName, PrognosisTestTheilsUStatisticAR1ResultDescription, new DoubleValue()));
|
---|
275 | ((DoubleValue)this[PrognosisTestTheilsUStatisticAR1ResultName].Value).Value = value;
|
---|
276 | }
|
---|
277 | }
|
---|
278 | public double PrognosisTrainingTheilsUStatisticMean {
|
---|
279 | get {
|
---|
280 | if (!ContainsKey(PrognosisTrainingTheilsUStatisticMeanResultName)) return double.NaN;
|
---|
281 | return ((DoubleValue)this[PrognosisTrainingTheilsUStatisticMeanResultName].Value).Value;
|
---|
282 | }
|
---|
283 | private set {
|
---|
284 | if (!ContainsKey(PrognosisTrainingTheilsUStatisticMeanResultName)) Add(new Result(PrognosisTrainingTheilsUStatisticMeanResultName, PrognosisTrainingTheilsUStatisticMeanResultDescription, new DoubleValue()));
|
---|
285 | ((DoubleValue)this[PrognosisTrainingTheilsUStatisticMeanResultName].Value).Value = value;
|
---|
286 | }
|
---|
287 | }
|
---|
288 | public double PrognosisTestTheilsUStatisticMean {
|
---|
289 | get {
|
---|
290 | if (!ContainsKey(PrognosisTestTheilsUStatisticMeanResultName)) return double.NaN;
|
---|
291 | return ((DoubleValue)this[PrognosisTestTheilsUStatisticMeanResultName].Value).Value;
|
---|
292 | }
|
---|
293 | private set {
|
---|
294 | if (!ContainsKey(PrognosisTestTheilsUStatisticMeanResultName)) Add(new Result(PrognosisTestTheilsUStatisticMeanResultName, PrognosisTestTheilsUStatisticMeanResultDescription, new DoubleValue()));
|
---|
295 | ((DoubleValue)this[PrognosisTestTheilsUStatisticMeanResultName].Value).Value = value;
|
---|
296 | }
|
---|
297 | }
|
---|
298 | #endregion
|
---|
299 |
|
---|
300 | [Storable]
|
---|
301 | private int trainingHorizon;
|
---|
302 | public int TrainingHorizon {
|
---|
303 | get { return trainingHorizon; }
|
---|
304 | set {
|
---|
305 | if (trainingHorizon != value) {
|
---|
306 | trainingHorizon = value;
|
---|
307 | OnTrainingHorizonChanged();
|
---|
308 | }
|
---|
309 | }
|
---|
310 | }
|
---|
311 |
|
---|
312 | [Storable]
|
---|
313 | private int testHorizon;
|
---|
314 | public int TestHorizon {
|
---|
315 | get { return testHorizon; }
|
---|
316 | set {
|
---|
317 | if (testHorizon != value) {
|
---|
318 | testHorizon = value;
|
---|
319 | OnTestHorizonChanged();
|
---|
320 | }
|
---|
321 | }
|
---|
322 | }
|
---|
323 |
|
---|
324 | private ITimeSeriesPrognosisSolution solution;
|
---|
325 | [Storable]
|
---|
326 | public ITimeSeriesPrognosisSolution Solution {
|
---|
327 | get { return solution; }
|
---|
328 | private set { solution = value; } //necessary for persistence
|
---|
329 | }
|
---|
330 |
|
---|
331 | [StorableConstructor]
|
---|
332 | public TimeSeriesPrognosisResults(bool deserializing) : base(deserializing) { }
|
---|
333 | protected TimeSeriesPrognosisResults(TimeSeriesPrognosisResults original, Cloner cloner)
|
---|
334 | : base(original, cloner) {
|
---|
335 | this.trainingHorizon = original.trainingHorizon;
|
---|
336 | this.testHorizon = original.testHorizon;
|
---|
337 | this.solution = cloner.Clone(original.solution);
|
---|
338 | }
|
---|
339 | public override IDeepCloneable Clone(Cloner cloner) {
|
---|
340 | return new TimeSeriesPrognosisResults(this, cloner);
|
---|
341 | }
|
---|
342 |
|
---|
343 | public TimeSeriesPrognosisResults(int trainingHorizon, int testHorizon, ITimeSeriesPrognosisSolution solution)
|
---|
344 | : base() {
|
---|
345 | this.trainingHorizon = trainingHorizon;
|
---|
346 | this.testHorizon = testHorizon;
|
---|
347 | this.solution = solution;
|
---|
348 | CalculateTrainingPrognosisResults();
|
---|
349 | CalculateTestPrognosisResults();
|
---|
350 | }
|
---|
351 |
|
---|
352 | #region events
|
---|
353 | public event EventHandler TrainingHorizonChanged;
|
---|
354 | protected virtual void OnTrainingHorizonChanged() {
|
---|
355 | CalculateTrainingPrognosisResults();
|
---|
356 | var handler = TrainingHorizonChanged;
|
---|
357 | if (handler != null) handler(this, EventArgs.Empty);
|
---|
358 | }
|
---|
359 |
|
---|
360 | public event EventHandler TestHorizonChanged;
|
---|
361 | protected virtual void OnTestHorizonChanged() {
|
---|
362 | CalculateTestPrognosisResults();
|
---|
363 | var handler = TestHorizonChanged;
|
---|
364 | if (handler != null) handler(this, EventArgs.Empty);
|
---|
365 | }
|
---|
366 | #endregion
|
---|
367 |
|
---|
368 | private void CalculateTrainingPrognosisResults() {
|
---|
369 | OnlineCalculatorError errorState;
|
---|
370 | var problemData = Solution.ProblemData;
|
---|
371 | if (!problemData.TrainingIndices.Any()) return;
|
---|
372 | var model = Solution.Model;
|
---|
373 | //mean model
|
---|
374 | double trainingMean = problemData.Dataset.GetDoubleValues(problemData.TargetVariable, problemData.TrainingIndices).Average();
|
---|
375 | var meanModel = new ConstantModel(trainingMean, problemData.TargetVariable);
|
---|
376 |
|
---|
377 | //AR1 model
|
---|
378 | double alpha, beta;
|
---|
379 | IEnumerable<double> trainingStartValues = problemData.Dataset.GetDoubleValues(problemData.TargetVariable, problemData.TrainingIndices.Select(r => r - 1).Where(r => r > 0)).ToList();
|
---|
380 | OnlineLinearScalingParameterCalculator.Calculate(problemData.Dataset.GetDoubleValues(problemData.TargetVariable, problemData.TrainingIndices.Where(x => x > 0)), trainingStartValues, out alpha, out beta, out errorState);
|
---|
381 | var AR1model = new TimeSeriesPrognosisAutoRegressiveModel(problemData.TargetVariable, new double[] { beta }, alpha);
|
---|
382 |
|
---|
383 | var trainingHorizions = problemData.TrainingIndices.Select(r => Math.Min(trainingHorizon, problemData.TrainingPartition.End - r)).ToList();
|
---|
384 | IEnumerable<IEnumerable<double>> trainingTargetValues = problemData.TrainingIndices.Zip(trainingHorizions, Enumerable.Range).Select(r => problemData.Dataset.GetDoubleValues(problemData.TargetVariable, r)).ToList();
|
---|
385 | IEnumerable<IEnumerable<double>> trainingEstimatedValues = model.GetPrognosedValues(problemData.Dataset, problemData.TrainingIndices, trainingHorizions).ToList();
|
---|
386 | IEnumerable<IEnumerable<double>> trainingMeanModelPredictions = meanModel.GetPrognosedValues(problemData.Dataset, problemData.TrainingIndices, trainingHorizions).ToList();
|
---|
387 | IEnumerable<IEnumerable<double>> trainingAR1ModelPredictions = AR1model.GetPrognosedValues(problemData.Dataset, problemData.TrainingIndices, trainingHorizions).ToList();
|
---|
388 |
|
---|
389 | IEnumerable<double> originalTrainingValues = trainingTargetValues.SelectMany(x => x).ToList();
|
---|
390 | IEnumerable<double> estimatedTrainingValues = trainingEstimatedValues.SelectMany(x => x).ToList();
|
---|
391 |
|
---|
392 | double trainingMSE = OnlineMeanSquaredErrorCalculator.Calculate(originalTrainingValues, estimatedTrainingValues, out errorState);
|
---|
393 | PrognosisTrainingMeanSquaredError = errorState == OnlineCalculatorError.None ? trainingMSE : double.NaN;
|
---|
394 | double trainingMAE = OnlineMeanAbsoluteErrorCalculator.Calculate(originalTrainingValues, estimatedTrainingValues, out errorState);
|
---|
395 | PrognosisTrainingMeanAbsoluteError = errorState == OnlineCalculatorError.None ? trainingMAE : double.NaN;
|
---|
396 | double trainingR = OnlinePearsonsRCalculator.Calculate(originalTrainingValues, estimatedTrainingValues, out errorState);
|
---|
397 | PrognosisTrainingRSquared = errorState == OnlineCalculatorError.None ? trainingR * trainingR : double.NaN;
|
---|
398 | double trainingRelError = OnlineMeanAbsolutePercentageErrorCalculator.Calculate(originalTrainingValues, estimatedTrainingValues, out errorState);
|
---|
399 | PrognosisTrainingRelativeError = errorState == OnlineCalculatorError.None ? trainingRelError : double.NaN;
|
---|
400 | double trainingNMSE = OnlineNormalizedMeanSquaredErrorCalculator.Calculate(originalTrainingValues, estimatedTrainingValues, out errorState);
|
---|
401 | PrognosisTrainingNormalizedMeanSquaredError = errorState == OnlineCalculatorError.None ? trainingNMSE : double.NaN;
|
---|
402 | double trainingME = OnlineMeanErrorCalculator.Calculate(originalTrainingValues, estimatedTrainingValues, out errorState);
|
---|
403 | PrognosisTrainingMeanError = errorState == OnlineCalculatorError.None ? trainingME : double.NaN;
|
---|
404 |
|
---|
405 | PrognosisTrainingDirectionalSymmetry = OnlineDirectionalSymmetryCalculator.Calculate(trainingStartValues, trainingTargetValues, trainingEstimatedValues, out errorState);
|
---|
406 | PrognosisTrainingDirectionalSymmetry = errorState == OnlineCalculatorError.None ? PrognosisTrainingDirectionalSymmetry : 0.0;
|
---|
407 | PrognosisTrainingWeightedDirectionalSymmetry = OnlineWeightedDirectionalSymmetryCalculator.Calculate(trainingStartValues, trainingTargetValues, trainingEstimatedValues, out errorState);
|
---|
408 | PrognosisTrainingWeightedDirectionalSymmetry = errorState == OnlineCalculatorError.None ? PrognosisTrainingWeightedDirectionalSymmetry : 0.0;
|
---|
409 | PrognosisTrainingTheilsUStatisticAR1 = OnlineTheilsUStatisticCalculator.Calculate(trainingStartValues, trainingTargetValues, trainingAR1ModelPredictions, trainingEstimatedValues, out errorState);
|
---|
410 | PrognosisTrainingTheilsUStatisticAR1 = errorState == OnlineCalculatorError.None ? PrognosisTrainingTheilsUStatisticAR1 : double.PositiveInfinity;
|
---|
411 | PrognosisTrainingTheilsUStatisticMean = OnlineTheilsUStatisticCalculator.Calculate(trainingStartValues, trainingTargetValues, trainingMeanModelPredictions, trainingEstimatedValues, out errorState);
|
---|
412 | PrognosisTrainingTheilsUStatisticMean = errorState == OnlineCalculatorError.None ? PrognosisTrainingTheilsUStatisticMean : double.PositiveInfinity;
|
---|
413 | }
|
---|
414 |
|
---|
415 | private void CalculateTestPrognosisResults() {
|
---|
416 | OnlineCalculatorError errorState;
|
---|
417 | var problemData = Solution.ProblemData;
|
---|
418 | if (!problemData.TestIndices.Any()) return;
|
---|
419 | var model = Solution.Model;
|
---|
420 | var testHorizions = problemData.TestIndices.Select(r => Math.Min(testHorizon, problemData.TestPartition.End - r)).ToList();
|
---|
421 | IEnumerable<IEnumerable<double>> testTargetValues = problemData.TestIndices.Zip(testHorizions, Enumerable.Range).Select(r => problemData.Dataset.GetDoubleValues(problemData.TargetVariable, r)).ToList();
|
---|
422 | IEnumerable<IEnumerable<double>> testEstimatedValues = model.GetPrognosedValues(problemData.Dataset, problemData.TestIndices, testHorizions).ToList();
|
---|
423 | IEnumerable<double> testStartValues = problemData.Dataset.GetDoubleValues(problemData.TargetVariable, problemData.TestIndices.Select(r => r - 1).Where(r => r > 0)).ToList();
|
---|
424 |
|
---|
425 | IEnumerable<double> originalTestValues = testTargetValues.SelectMany(x => x).ToList();
|
---|
426 | IEnumerable<double> estimatedTestValues = testEstimatedValues.SelectMany(x => x).ToList();
|
---|
427 |
|
---|
428 | double testMSE = OnlineMeanSquaredErrorCalculator.Calculate(originalTestValues, estimatedTestValues, out errorState);
|
---|
429 | PrognosisTestMeanSquaredError = errorState == OnlineCalculatorError.None ? testMSE : double.NaN;
|
---|
430 | double testMAE = OnlineMeanAbsoluteErrorCalculator.Calculate(originalTestValues, estimatedTestValues, out errorState);
|
---|
431 | PrognosisTestMeanAbsoluteError = errorState == OnlineCalculatorError.None ? testMAE : double.NaN;
|
---|
432 | double testR = OnlinePearsonsRCalculator.Calculate(originalTestValues, estimatedTestValues, out errorState);
|
---|
433 | PrognosisTestRSquared = errorState == OnlineCalculatorError.None ? testR * testR : double.NaN;
|
---|
434 | double testRelError = OnlineMeanAbsolutePercentageErrorCalculator.Calculate(originalTestValues, estimatedTestValues, out errorState);
|
---|
435 | PrognosisTestRelativeError = errorState == OnlineCalculatorError.None ? testRelError : double.NaN;
|
---|
436 | double testNMSE = OnlineNormalizedMeanSquaredErrorCalculator.Calculate(originalTestValues, estimatedTestValues, out errorState);
|
---|
437 | PrognosisTestNormalizedMeanSquaredError = errorState == OnlineCalculatorError.None ? testNMSE : double.NaN;
|
---|
438 | double testME = OnlineMeanErrorCalculator.Calculate(originalTestValues, estimatedTestValues, out errorState);
|
---|
439 | PrognosisTestMeanError = errorState == OnlineCalculatorError.None ? testME : double.NaN;
|
---|
440 |
|
---|
441 | PrognosisTestDirectionalSymmetry = OnlineDirectionalSymmetryCalculator.Calculate(testStartValues, testTargetValues, testEstimatedValues, out errorState);
|
---|
442 | PrognosisTestDirectionalSymmetry = errorState == OnlineCalculatorError.None ? PrognosisTestDirectionalSymmetry : 0.0;
|
---|
443 | PrognosisTestWeightedDirectionalSymmetry = OnlineWeightedDirectionalSymmetryCalculator.Calculate(testStartValues, testTargetValues, testEstimatedValues, out errorState);
|
---|
444 | PrognosisTestWeightedDirectionalSymmetry = errorState == OnlineCalculatorError.None ? PrognosisTestWeightedDirectionalSymmetry : 0.0;
|
---|
445 |
|
---|
446 |
|
---|
447 | if (problemData.TrainingIndices.Any()) {
|
---|
448 | //mean model
|
---|
449 | double trainingMean = problemData.Dataset.GetDoubleValues(problemData.TargetVariable, problemData.TrainingIndices).Average();
|
---|
450 | var meanModel = new ConstantModel(trainingMean, problemData.TargetVariable);
|
---|
451 |
|
---|
452 | //AR1 model
|
---|
453 | double alpha, beta;
|
---|
454 | IEnumerable<double> trainingStartValues = problemData.Dataset.GetDoubleValues(problemData.TargetVariable, problemData.TrainingIndices.Select(r => r - 1).Where(r => r > 0)).ToList();
|
---|
455 | OnlineLinearScalingParameterCalculator.Calculate(problemData.Dataset.GetDoubleValues(problemData.TargetVariable, problemData.TrainingIndices.Where(x => x > 0)), trainingStartValues, out alpha, out beta, out errorState);
|
---|
456 | var AR1model = new TimeSeriesPrognosisAutoRegressiveModel(problemData.TargetVariable, new double[] { beta }, alpha);
|
---|
457 |
|
---|
458 | IEnumerable<IEnumerable<double>> testMeanModelPredictions = meanModel.GetPrognosedValues(problemData.Dataset, problemData.TestIndices, testHorizions).ToList();
|
---|
459 | IEnumerable<IEnumerable<double>> testAR1ModelPredictions = AR1model.GetPrognosedValues(problemData.Dataset, problemData.TestIndices, testHorizions).ToList();
|
---|
460 |
|
---|
461 | PrognosisTestTheilsUStatisticAR1 = OnlineTheilsUStatisticCalculator.Calculate(testStartValues, testTargetValues, testAR1ModelPredictions, testEstimatedValues, out errorState);
|
---|
462 | PrognosisTestTheilsUStatisticAR1 = errorState == OnlineCalculatorError.None ? PrognosisTestTheilsUStatisticAR1 : double.PositiveInfinity;
|
---|
463 | PrognosisTestTheilsUStatisticMean = OnlineTheilsUStatisticCalculator.Calculate(testStartValues, testTargetValues, testMeanModelPredictions, testEstimatedValues, out errorState);
|
---|
464 | PrognosisTestTheilsUStatisticMean = errorState == OnlineCalculatorError.None ? PrognosisTestTheilsUStatisticMean : double.PositiveInfinity;
|
---|
465 | }
|
---|
466 | }
|
---|
467 | }
|
---|
468 | }
|
---|