num4anova 0.1.2-java → 0.2.1-java

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (5) hide show
  1. checksums.yaml +4 -4
  2. data/CHANGELOG.md +8 -0
  3. data/ext/num4anova/Ancova.java +125 -125
  4. data/lib/num4anova.rb +74 -81
  5. metadata +2 -2
checksums.yaml CHANGED
@@ -1,7 +1,7 @@
1
1
  ---
2
2
  SHA256:
3
- metadata.gz: ab6d4ccfb1be143b1d3884f5946a053b48f64176c2c629586798db9c0a286e28
4
- data.tar.gz: 74b4bb07862055a7e0be83a9e3c0de07ff9272715b70e4c9d7f84f7134329f1c
3
+ metadata.gz: cee991016acc5e70a9d573a3c854a4c59e0f7d743e68873fe17b0a7e9fd105b3
4
+ data.tar.gz: f1c465691191a15cc751956129a4bc8470ecec63a14d7cea8bd04faf47df61e6
5
5
  SHA512:
6
- metadata.gz: ab2c722af19199df5a51ac6891c309ed322ba4c4d576bc7bef04afbdf99a8ce770f6792aa6a3bb867d621d7ba7ffa855784f88b0eb08c509fd8aefc2d5bab837
7
- data.tar.gz: 608304d1297cdb9282376790a00cfe8dd891ec00e254a79e7e566e2bac9385b1ce3489fdab10db9bd9f1b733faa912379371d25dbb026e36f87df25ed4360239
6
+ metadata.gz: f8608fc8d57d8a1da8fa90b54463d0e7a8b8a43cf2f2fe020b7584649a397d952b2d441730af24c9dbcfebae90c73b0953be29934055eca70f408b1e11cec0d0
7
+ data.tar.gz: d164367e6427c6339b89a6be3ad8e9b43cd7ea6c1a42c8c3c6b3eea0fbe971e723761f7904abe386e9437396a19b3030bcd3d1f60e6abda276ff0e1c8d1e2d88
data/CHANGELOG.md CHANGED
@@ -2,6 +2,14 @@
2
2
 
3
3
  ## Unreleased
4
4
 
5
+ ## [0.2.1] - 2024-06-17
6
+ ### chg
7
+ - chg Num4AncovaLib.
8
+
9
+ ## [0.1.3] - 2024-05-28
10
+ ### chg
11
+ - chg to_a in create_oneway.
12
+
5
13
  ## [0.1.2] - 2024-05-27
6
14
  ### add
7
15
  - add function of create_oneway in TwoWayLayoutLib.
data/ext/num4anova/Ancova.java CHANGED
@@ -6,38 +6,38 @@ public class Ancova {
6
6
  public static Ancova getInstance() {
7
7
  return ancova;
8
8
  }
9
- public boolean parallelTest(double[][][] xi, double a) {
9
+ public boolean parallelTest(double yi[][], double[][] xi, double a) {
10
10
  HypothesisTest hypoth = new Parallettest();
11
11
 
12
- double statistic = hypoth.calcTestStatistic(xi);
12
+ double statistic = hypoth.calcTestStatistic(yi, xi);
13
13
  return hypoth.executeTest(statistic, a);
14
14
  }
15
- public boolean significanceTest(double[][][] xi, double a) {
15
+ public boolean significanceTest(double yi[][], double[][] xi, double a) {
16
16
  HypothesisTest hypoth = new SignificanceTest();
17
17
 
18
- double statistic = hypoth.calcTestStatistic(xi);
18
+ double statistic = hypoth.calcTestStatistic(yi, xi);
19
19
  return hypoth.executeTest(statistic, a);
20
20
  }
21
- public boolean differenceTest(double[][][] xi, double a) {
21
+ public boolean differenceTest(double yi[][], double[][] xi, double a) {
22
22
  HypothesisTest hypoth = new DifferenceTest();
23
23
 
24
- double statistic = hypoth.calcTestStatistic(xi);
24
+ double statistic = hypoth.calcTestStatistic(yi, xi);
25
25
  return hypoth.executeTest(statistic, a);
26
26
  }
27
- public Interval[] intervalEstim(double[][][] xi, double a) {
27
+ public Interval[] intervalEstim(double yi[][], double[][] xi, double a) {
28
28
  Estim estim = new IntervalEstim();
29
29
 
30
- return estim.calcInterval(xi, a);
30
+ return estim.calcInterval(yi, xi, a);
31
31
  }
32
32
  /*********************************/
33
33
  /* interface define */
34
34
  /*********************************/
35
35
  private interface HypothesisTest {
36
- double calcTestStatistic(double[][][] xi);
36
+ double calcTestStatistic(double yi[][], double[][] xi);
37
37
  boolean executeTest(double statistic, double a);
38
38
  }
39
39
  private interface Estim {
40
- Interval[] calcInterval(double[][][] xi, double a);
40
+ Interval[] calcInterval(double yi[][], double[][] xi, double a);
41
41
  }
42
42
  /*********************************/
43
43
  /* class define */
@@ -53,7 +53,7 @@ public class Ancova {
53
53
  public double getMax() { return this.max; }
54
54
  }
55
55
  private class RegressionLine {
56
- protected int calcSumn(double[][][] xi) {
56
+ protected int calcSumn(double[][] xi) {
57
57
  int sum = 0;
58
58
 
59
59
  for(int i = 0; i < xi.length; i++) {
@@ -62,69 +62,69 @@ public class Ancova {
62
62
  return sum;
63
63
  }
64
64
  // 全変動
65
- private double calcSty(double[][][] xi, int sumn) {
66
- double sumy2 = calcSumy2(xi);
67
- double sumy = calcSumy(xi);
65
+ private double calcSty(double yi[][], int sumn) {
66
+ double sumy2 = calcSumy2(yi);
67
+ double sumy = calcSumy(yi);
68
68
 
69
69
  return sumy2 - sumy*sumy / sumn;
70
70
  }
71
- private double calcStx(double[][][] xi, int sumn) {
71
+ private double calcStx(double[][] xi, int sumn) {
72
72
  double sumx2 = calcSumx2(xi);
73
73
  double sumx = calcSumx(xi);
74
74
 
75
75
  return sumx2 - sumx*sumx / sumn;
76
76
  }
77
- private double calcStyx(double[][][] xi, int sumn) {
77
+ private double calcStyx(double yi[][], double[][] xi, int sumn) {
78
78
  double sumx = calcSumx(xi);
79
- double sumy = calcSumy(xi);
80
- double sumyx = calcSumyx(xi);
79
+ double sumy = calcSumy(yi);
80
+ double sumyx = calcSumyx(yi, xi);
81
81
 
82
82
  return sumyx - sumy*sumx / sumn;
83
83
  }
84
84
  // 水準間変動
85
- private double calcSay(double[][][] xi, int sumn) {
86
- double sumy = calcSumy(xi);
85
+ private double calcSay(double yi[][], int sumn) {
86
+ double sumy = calcSumy(yi);
87
87
 
88
- return calcSumay(xi) - sumy*sumy / sumn;
88
+ return calcSumay(yi) - sumy*sumy / sumn;
89
89
  }
90
- private double calcSax(double[][][] xi, int sumn) {
90
+ private double calcSax(double[][] xi, int sumn) {
91
91
  double sumx = calcSumx(xi);
92
92
 
93
93
  return calcSumax(xi) - sumx*sumx / sumn;
94
94
  }
95
- private double calcSayx(double[][][] xi, int sumn) {
95
+ private double calcSayx(double yi[][], double[][] xi, int sumn) {
96
96
  double sumx = calcSumx(xi);
97
- double sumy = calcSumy(xi);
97
+ double sumy = calcSumy(yi);
98
98
 
99
- return calcSumayx(xi) - sumy*sumx / sumn;
99
+ return calcSumayx(yi, xi) - sumy*sumx / sumn;
100
100
  }
101
101
  // 水準内変動
102
- protected double calcSex(double[][][] xi, int sumn) {
102
+ protected double calcSex(double[][] xi, int sumn) {
103
103
  return calcStx(xi, sumn) - calcSax(xi, sumn);
104
104
  }
105
- protected double calcSey(double[][][] xi, int sumn) {
106
- return calcSty(xi, sumn) - calcSay(xi, sumn);
105
+ protected double calcSey(double yi[][], int sumn) {
106
+ return calcSty(yi, sumn) - calcSay(yi, sumn);
107
107
  }
108
- protected double calcSeyx(double[][][] xi, int sumn) {
109
- return calcStyx(xi, sumn) - calcSayx(xi, sumn);
108
+ protected double calcSeyx(double yi[][], double[][] xi, int sumn) {
109
+ return calcStyx(yi, xi, sumn) - calcSayx(yi, xi, sumn);
110
110
  }
111
111
 
112
112
  // 平行性の検定
113
- protected double calcbx(double[][][] xi) {
113
+ protected double calcbx(double yi[][], double[][] xi) {
114
114
  double sum = 0.0;
115
115
 
116
- for (int i = 0; i < xi.length; i++) {
117
- int n = xi[i].length;
116
+ for (int i = 0; i < yi.length; i++) {
117
+ int n = yi[i].length;
118
118
  double sumx = 0.0;
119
119
  double sumy = 0.0;
120
120
  double sumyx = 0.0;
121
121
  double sumx2 = 0.0;
122
122
  for (int j = 0; j < n; j++) {
123
- sumx += xi[i][j][1];
124
- sumy += xi[i][j][0];
125
- sumyx += xi[i][j][1] * xi[i][j][0];
123
+ sumx += xi[i][j];
124
+ sumy += yi[i][j];
125
+ sumyx += xi[i][j] * yi[i][j];
126
126
 
127
- sumx2 += xi[i][j][1] * xi[i][j][1];
127
+ sumx2 += xi[i][j] * xi[i][j];
128
128
 
129
129
  }
130
130
  double wki = n * sumyx - sumy * sumx;
@@ -135,12 +135,12 @@ public class Ancova {
135
135
  return sum;
136
136
  }
137
137
  // 差の検定
138
- protected double calcSa(double[][][] xi, int sumn) {
139
- double sumty = calcSty(xi, sumn);
140
- double sumtyx = calcStyx(xi, sumn);
138
+ protected double calcSa(double yi[][], double[][] xi, int sumn) {
139
+ double sumty = calcSty(yi, sumn);
140
+ double sumtyx = calcStyx(yi, xi, sumn);
141
141
  double sumtx = calcStx(xi, sumn);
142
- double sumey = calcSey(xi, sumn);
143
- double sumeyx = calcSeyx(xi, sumn);
142
+ double sumey = calcSey(yi, sumn);
143
+ double sumeyx = calcSeyx(yi, xi, sumn);
144
144
  double sumex = calcSex(xi, sumn);
145
145
 
146
146
  return (sumty - sumtyx * sumtyx / sumtx)
@@ -148,90 +148,90 @@ public class Ancova {
148
148
  }
149
149
 
150
150
  // ETC
151
- private double calcSumay(double[][][] xi) {
151
+ private double calcSumay(double yi[][]) {
152
152
  double sum = 0.0;
153
153
 
154
- for (int i = 0; i < xi.length; i++) {
154
+ for (int i = 0; i < yi.length; i++) {
155
155
  double sumyi = 0.0;
156
- for (int j = 0; j < xi[i].length; j++) {
157
- sumyi += xi[i][j][0];
156
+ for (int j = 0; j < yi[i].length; j++) {
157
+ sumyi += yi[i][j];
158
158
  }
159
- sum += sumyi * sumyi / xi[i].length;
159
+ sum += sumyi * sumyi / yi[i].length;
160
160
  }
161
161
  return sum;
162
162
  }
163
- private double calcSumy2(double[][][] xi) {
163
+ private double calcSumy2(double yi[][]) {
164
164
  double sum = 0.0;
165
165
 
166
- for (int i = 0; i < xi.length; i++) {
167
- for (int j = 0; j < xi[i].length; j++) {
168
- sum += xi[i][j][0] * xi[i][j][0];
166
+ for (int i = 0; i < yi.length; i++) {
167
+ for (int j = 0; j < yi[i].length; j++) {
168
+ sum += yi[i][j] * yi[i][j];
169
169
  }
170
170
  }
171
171
  return sum;
172
172
  }
173
- private double calcSumx(double[][][] xi) {
173
+ private double calcSumx(double[][] xi) {
174
174
  double sum = 0.0;
175
175
 
176
176
  for (int i = 0; i < xi.length; i++) {
177
177
  for (int j = 0; j < xi[i].length; j++) {
178
- sum += xi[i][j][1];
178
+ sum += xi[i][j];
179
179
  }
180
180
  }
181
181
  return sum;
182
182
  }
183
- protected double calcSumx2(double[][][] xi) {
183
+ protected double calcSumx2(double[][] xi) {
184
184
  double sum = 0.0;
185
185
 
186
186
  for (int i = 0; i < xi.length; i++) {
187
187
  for (int j = 0; j < xi[i].length; j++) {
188
- sum += xi[i][j][1] * xi[i][j][1];
188
+ sum += xi[i][j] * xi[i][j];
189
189
  }
190
190
  }
191
191
  return sum;
192
192
  }
193
- protected double calcSumyx(double[][][] xi) {
193
+ protected double calcSumyx(double yi[][], double[][] xi) {
194
194
  double sum = 0.0;
195
195
 
196
- for (int i = 0; i < xi.length; i++) {
197
- for (int j = 0; j < xi[i].length; j++) {
198
- sum += xi[i][j][0] * xi[i][j][1];
196
+ for (int i = 0; i < yi.length; i++) {
197
+ for (int j = 0; j < yi[i].length; j++) {
198
+ sum += yi[i][j] * xi[i][j];
199
199
  }
200
200
  }
201
201
  return sum;
202
202
  }
203
- private double calcSumayx(double[][][] xi) {
203
+ private double calcSumayx(double yi[][], double[][] xi) {
204
204
  double sum = 0.0;
205
205
 
206
- for (int i = 0; i < xi.length; i++) {
206
+ for (int i = 0; i < yi.length; i++) {
207
207
  double sumxi = 0.0;
208
208
  double sumyi = 0.0;
209
- for (int j = 0; j < xi[i].length; j++) {
210
- sumxi += xi[i][j][1];
211
- sumyi += xi[i][j][0];
209
+ for (int j = 0; j < yi[i].length; j++) {
210
+ sumxi += xi[i][j];
211
+ sumyi += yi[i][j];
212
212
  }
213
213
  sum += sumxi * sumyi / xi[i].length;
214
214
  }
215
215
  return sum;
216
216
  }
217
- private double calcSumax(double[][][] xi) {
217
+ private double calcSumax(double[][] xi) {
218
218
  double sum = 0.0;
219
219
 
220
220
  for (int i = 0; i < xi.length; i++) {
221
221
  double sumxi = 0.0;
222
222
  for (int j = 0; j < xi[i].length; j++) {
223
- sumxi += xi[i][j][1];
223
+ sumxi += xi[i][j];
224
224
  }
225
225
  sum += sumxi * sumxi / xi[i].length;
226
226
  }
227
227
  return sum;
228
228
  }
229
- private double calcSumy(double[][][] xi) {
229
+ private double calcSumy(double yi[][]) {
230
230
  double sum = 0.0;
231
231
 
232
- for (int i = 0; i < xi.length; i++) {
233
- for (int j = 0; j < xi[i].length; j++) {
234
- sum += xi[i][j][0];
232
+ for (int i = 0; i < yi.length; i++) {
233
+ for (int j = 0; j < yi[i].length; j++) {
234
+ sum += yi[i][j];
235
235
  }
236
236
  }
237
237
  return sum;
@@ -242,32 +242,32 @@ public class Ancova {
242
242
  private class Parallettest extends RegressionLine implements HypothesisTest {
243
243
  private int n = 0;
244
244
  private int m = 0;
245
- public double calcTestStatistic(double[][][] xi) {
245
+ public double calcTestStatistic(double yi[][], double[][] xi) {
246
246
  int sumn = calcSumn(xi);
247
- n = xi.length - 1;
248
- m = sumn - 2 * xi.length;
247
+ n = yi.length - 1;
248
+ m = sumn - 2 * yi.length;
249
249
 
250
- double vnp = calcVnp(xi, sumn);
251
- double ve2 = calcVe2(xi, sumn);
250
+ double vnp = calcVnp(yi, xi, sumn);
251
+ double ve2 = calcVe2(yi, xi, sumn);
252
252
 
253
253
  return vnp / ve2;
254
254
  }
255
- private double calcVnp(double[][][] xi, int sumn){
256
- return calcSnp(xi, sumn) / n;
255
+ private double calcVnp(double yi[][], double[][] xi, int sumn){
256
+ return calcSnp(yi, xi, sumn) / n;
257
257
  }
258
- private double calcSnp(double[][][] xi, int sumn) {
259
- double sumbx = calcbx(xi);
260
- double sumeyx = calcSeyx(xi, sumn);
258
+ private double calcSnp(double yi[][], double[][] xi, int sumn) {
259
+ double sumbx = calcbx(yi, xi);
260
+ double sumeyx = calcSeyx(yi, xi, sumn);
261
261
  double sumex = calcSex(xi, sumn);
262
262
 
263
263
  return sumbx - sumeyx * sumeyx / sumex;
264
264
  }
265
- private double calcVe2(double[][][] xi, int sumn) {
266
- return calcSe2(xi, sumn) / m;
265
+ private double calcVe2(double yi[][], double[][] xi, int sumn) {
266
+ return calcSe2(yi, xi, sumn) / m;
267
267
  }
268
- private double calcSe2(double[][][] xi, int sumn) {
269
- double sumey = calcSey(xi, sumn);
270
- double sumbx = calcbx(xi);
268
+ private double calcSe2(double yi[][], double[][] xi, int sumn) {
269
+ double sumey = calcSey(yi, sumn);
270
+ double sumbx = calcbx(yi, xi);
271
271
 
272
272
  return sumey - sumbx;
273
273
  }
@@ -282,13 +282,13 @@ public class Ancova {
282
282
  private class SignificanceTest extends RegressionLine implements HypothesisTest {
283
283
  private int n = 0;
284
284
  private int m = 0;
285
- public double calcTestStatistic(double[][][] xi) {
285
+ public double calcTestStatistic(double yi[][], double[][] xi) {
286
286
  int sumn = calcSumn(xi);
287
287
  n = 1;
288
288
  m = sumn - xi.length - 1;
289
289
 
290
- double vr = calcVr(xi, sumn);
291
- double ve = calcVe(xi, sumn);
290
+ double vr = calcVr(yi, xi, sumn);
291
+ double ve = calcVe(yi, xi, sumn);
292
292
 
293
293
  return vr / ve;
294
294
  }
@@ -298,16 +298,16 @@ public class Ancova {
298
298
 
299
299
  return (statistic >= f) ? true : false;
300
300
  }
301
- private double calcVr(double[][][] xi, int sumn) {
302
- double sumeyx = calcSeyx(xi, sumn);
301
+ private double calcVr(double yi[][], double[][] xi, int sumn) {
302
+ double sumeyx = calcSeyx(yi, xi, sumn);
303
303
  double sumex = calcSex(xi, sumn);
304
304
 
305
305
  return (sumeyx * sumeyx) / sumex;
306
306
  }
307
- private double calcVe(double[][][] xi, int sumn) {
308
- double sumey = calcSey(xi, sumn);
307
+ private double calcVe(double yi[][], double[][] xi, int sumn) {
308
+ double sumey = calcSey(yi, sumn);
309
309
  double sumex = calcSex(xi, sumn);
310
- double sumeyx = calcSeyx(xi, sumn);
310
+ double sumeyx = calcSeyx(yi, xi, sumn);
311
311
 
312
312
  return (sumey * sumex - sumeyx * sumeyx) / (m * sumex);
313
313
  }
@@ -316,13 +316,13 @@ public class Ancova {
316
316
  private class DifferenceTest extends RegressionLine implements HypothesisTest {
317
317
  private int n = 0;
318
318
  private int m = 0;
319
- public double calcTestStatistic(double[][][] xi) {
319
+ public double calcTestStatistic(double yi[][], double[][] xi) {
320
320
  int sumn = calcSumn(xi);
321
- n = xi.length - 1;
322
- m = sumn - xi.length - 1;
321
+ n = yi.length - 1;
322
+ m = sumn - yi.length - 1;
323
323
 
324
- double va = calcSa(xi, sumn) / n;
325
- double ve = calcVe(xi, sumn);
324
+ double va = calcSa(yi, xi, sumn) / n;
325
+ double ve = calcVe(yi, xi, sumn);
326
326
 
327
327
  return va / ve;
328
328
  }
@@ -332,10 +332,10 @@ public class Ancova {
332
332
 
333
333
  return (statistic >= f) ? true : false;
334
334
  }
335
- private double calcVe(double[][][] xi, int sumn) {
336
- double sumey = calcSey(xi, sumn);
335
+ private double calcVe(double yi[][], double[][] xi, int sumn) {
336
+ double sumey = calcSey(yi, sumn);
337
337
  double sumex = calcSex(xi, sumn);
338
- double sumeyx = calcSeyx(xi, sumn);
338
+ double sumeyx = calcSeyx(yi, xi, sumn);
339
339
 
340
340
  return (sumey * sumex - sumeyx * sumeyx) / (m * sumex);
341
341
  }
@@ -346,16 +346,16 @@ public class Ancova {
346
346
  private int n = 0;
347
347
  private int[] ni = null;
348
348
  private double sumex = 0.0;
349
- public Interval[] calcInterval(double[][][] xi, double a) {
349
+ public Interval[] calcInterval(double yi[][], double[][] xi, double a) {
350
350
  Interval[] ret = new Interval[xi.length];
351
- ni = calcNi(xi);
351
+ ni = calcNi(yi);
352
352
  int sumn = calcSumn(xi);
353
353
  n = sumn - xi.length - 1;
354
354
  sumex = calcSex(xi, sumn);
355
- double ve = calcVe(xi, sumn);
356
- double b = calcB(xi, sumn);
355
+ double ve = calcVe(yi, xi, sumn);
356
+ double b = calcB(yi, xi, sumn);
357
357
 
358
- double[] meanyi = calcMeanyi(xi);
358
+ double[] meanyi = calcMeanyi(yi);
359
359
  double[] meanxi = calcMeanxi(xi);
360
360
  double meanx = calcMeanx(xi);
361
361
 
@@ -372,56 +372,56 @@ public class Ancova {
372
372
 
373
373
  return ret;
374
374
  }
375
- private int[] calcNi(double[][][] xi) {
376
- int[] ni = new int[xi.length];
375
+ private int[] calcNi(double[][] yi) {
376
+ int[] ni = new int[yi.length];
377
377
 
378
- for(int i = 0; i < xi.length; i++) {
379
- ni[i] = xi[i].length;
378
+ for(int i = 0; i < yi.length; i++) {
379
+ ni[i] = yi[i].length;
380
380
  }
381
381
  return ni;
382
382
  }
383
- private double calcVe(double[][][] xi, int sumn) {
384
- double sumey = calcSey(xi, sumn);
385
- double sumeyx = calcSeyx(xi, sumn);
383
+ private double calcVe(double[][] yi, double[][] xi, int sumn) {
384
+ double sumey = calcSey(yi, sumn);
385
+ double sumeyx = calcSeyx(yi, xi, sumn);
386
386
 
387
387
  return (sumey * sumex - sumeyx * sumeyx) / (n * sumex);
388
388
  }
389
- private double calcB(double[][][] xi, int sumn) {
390
- double sumeyx = calcSeyx(xi, sumn);
389
+ private double calcB(double[][] yi, double[][] xi, int sumn) {
390
+ double sumeyx = calcSeyx(yi, xi, sumn);
391
391
  double sex = calcSex(xi, sumn);
392
392
 
393
393
  return sumeyx / sex;
394
394
  }
395
- private double[] calcMeanyi(double[][][] xi) {
396
- double[] meanyi = new double[xi.length];
395
+ private double[] calcMeanyi(double[][] yi) {
396
+ double[] meanyi = new double[yi.length];
397
397
 
398
- for(int i = 0; i < xi.length; i++) {
398
+ for(int i = 0; i < yi.length; i++) {
399
399
  double sum = 0.0;
400
- for(int j = 0; j < xi[i].length; j++) {
401
- sum += xi[i][j][0];
400
+ for(int j = 0; j < yi[i].length; j++) {
401
+ sum += yi[i][j];
402
402
  }
403
- meanyi[i] = sum / xi[i].length;
403
+ meanyi[i] = sum / yi[i].length;
404
404
  }
405
405
  return meanyi;
406
406
  }
407
- private double[] calcMeanxi(double[][][] xi) {
407
+ private double[] calcMeanxi(double[][] xi) {
408
408
  double[] meanxi = new double[xi.length];
409
409
 
410
410
  for(int i = 0; i < xi.length; i++) {
411
411
  double sum = 0.0;
412
412
  for(int j = 0; j < xi[i].length; j++) {
413
- sum += xi[i][j][1];
413
+ sum += xi[i][j];
414
414
  }
415
415
  meanxi[i] = sum / xi[i].length;
416
416
  }
417
417
  return meanxi;
418
418
  }
419
- private double calcMeanx(double[][][] xi) {
419
+ private double calcMeanx(double[][] xi) {
420
420
  double sum = 0.0;
421
421
  double n = 0;
422
422
  for(int i = 0; i < xi.length; i++) {
423
423
  for(int j = 0; j < xi[i].length; j++) {
424
- sum += xi[i][j][1];
424
+ sum += xi[i][j];
425
425
  n++;
426
426
  }
427
427
  }
data/lib/num4anova.rb CHANGED
@@ -295,7 +295,8 @@ module Num4AnovaLib
295
295
  # [28.0, 31.2, 15.8],
296
296
  # ]
297
297
  def create_oneway(xij)
298
- return @twoWay.createOneWay(xij.to_java(Java::double[][]))
298
+ ret = @twoWay.createOneWay(xij.to_java(Java::double[][]))
299
+ return ret.to_a
299
300
  end
300
301
  end
301
302
  # 共分散分析
@@ -305,129 +306,121 @@ module Num4AnovaLib
305
306
  end
306
307
  # 回帰直線の平行性検定
307
308
  #
308
- # @overload parallel_test(xi, a)
309
- # @param [array] xi データ(double[][][])
309
+ # @overload parallel_test(yi, xi, a)
310
+ # @param [array] yi データ(double[][])
311
+ # @param [array] xi データ(double[][])
310
312
  # @param [double] a 有意水準
311
313
  # @return [boolean] 検定結果(boolean true:棄却域内 false:棄却域外)
312
314
  # @example
315
+ # yi = [
316
+ # [3, 5, 3],
317
+ # [3, 3, 8],
318
+ # [2, 2, 2],
319
+ # [3, 4, 2],
320
+ # [1, 2, 0],
321
+ # ]
313
322
  # xi = [
314
- # [
315
- # [3,35], [5,38], [3,39],
316
- # ],
317
- # [
318
- # [3,36], [3,39], [8,54],
319
- # ],
320
- # [
321
- # [2,40], [2,45], [2,39],
322
- # ],
323
- # [
324
- # [3,47], [4,52], [2,48],
325
- # ],
326
- # [
327
- # [1,64], [2,80], [0,70],
328
- # ],
323
+ # [35, 38, 39],
324
+ # [36, 39, 54],
325
+ # [40, 45, 39],
326
+ # [47, 52, 48],
327
+ # [64, 80, 70],
329
328
  # ]
330
329
  # ancova = Num4AnovaLib::Num4AncovaLib.new
331
- # ancova.parallel_test(xi, 0.05)
330
+ # ancova.parallel_test(yi, xi, 0.05)
332
331
  # => false
333
- def parallel_test(xi, a)
334
- @ancova.parallelTest(xi.to_java(Java::double[][]), a)
332
+ def parallel_test(yi, xi, a)
333
+ @ancova.parallelTest(yi.to_java(Java::double[]), xi.to_java(Java::double[]), a)
335
334
  end
336
335
  # 回帰直線の有意性検定
337
336
  #
338
- # @overload significance_test(xi, a)
339
- # @param [array] xi データ(double[][][])
337
+ # @overload significance_test(yi, xi, a)
338
+ # @param [array] yi データ(double[][])
339
+ # @param [array] xi データ(double[][])
340
340
  # @param [double] a 有意水準
341
341
  # @return [boolean] 検定結果(boolean true:棄却域内 false:棄却域外)
342
342
  # @example
343
+ # yi = [
344
+ # [3, 5, 3],
345
+ # [3, 3, 8],
346
+ # [2, 2, 2],
347
+ # [3, 4, 2],
348
+ # [1, 2, 0],
349
+ # ]
343
350
  # xi = [
344
- # [
345
- # [3,35], [5,38], [3,39],
346
- # ],
347
- # [
348
- # [3,36], [3,39], [8,54],
349
- # ],
350
- # [
351
- # [2,40], [2,45], [2,39],
352
- # ],
353
- # [
354
- # [3,47], [4,52], [2,48],
355
- # ],
356
- # [
357
- # [1,64], [2,80], [0,70],
358
- # ],
351
+ # [35, 38, 39],
352
+ # [36, 39, 54],
353
+ # [40, 45, 39],
354
+ # [47, 52, 48],
355
+ # [64, 80, 70],
359
356
  # ]
360
357
  # ancova = Num4AnovaLib::Num4AncovaLib.new
361
- # ancova.significance_test(xi, 0.05)
358
+ # ancova.significance_test(yi, xi, 0.05)
362
359
  # => true
363
- def significance_test(xi, a)
364
- @ancova.significanceTest(xi.to_java(Java::double[][]), a)
360
+ def significance_test(yi, xi, a)
361
+ @ancova.significanceTest(yi.to_java(Java::double[]), xi.to_java(Java::double[]), a)
365
362
  end
366
363
  # 水準間の差の検定
367
364
  #
368
- # @overload difference_test(xi, a)
369
- # @param [array] xi データ(double[][][])
365
+ # @overload difference_test(yi, xi, a)
366
+ # @param [array] yi データ(double[][])
367
+ # @param [array] xi データ(double[][])
370
368
  # @param [double] a 有意水準
371
369
  # @return [boolean] 検定結果(boolean true:棄却域内 false:棄却域外)
372
370
  # @example
371
+ # yi = [
372
+ # [3, 5, 3],
373
+ # [3, 3, 8],
374
+ # [2, 2, 2],
375
+ # [3, 4, 2],
376
+ # [1, 2, 0],
377
+ # ]
373
378
  # xi = [
374
- # [
375
- # [3,35], [5,38], [3,39],
376
- # ],
377
- # [
378
- # [3,36], [3,39], [8,54],
379
- # ],
380
- # [
381
- # [2,40], [2,45], [2,39],
382
- # ],
383
- # [
384
- # [3,47], [4,52], [2,48],
385
- # ],
386
- # [
387
- # [1,64], [2,80], [0,70],
388
- # ],
379
+ # [35, 38, 39],
380
+ # [36, 39, 54],
381
+ # [40, 45, 39],
382
+ # [47, 52, 48],
383
+ # [64, 80, 70],
389
384
  # ]
390
385
  # ancova = Num4AnovaLib::Num4AncovaLib.new
391
- # ancova.difference_test(xi, 0.05)
386
+ # ancova.difference_test(yi, xi, 0.05)
392
387
  # => true
393
- def difference_test(xi, a)
394
- @ancova.differenceTest(xi.to_java(Java::double[][]), a)
388
+ def difference_test(yi, xi, a)
389
+ @ancova.differenceTest(yi.to_java(Java::double[]), xi.to_java(Java::double[]), a)
395
390
  end
396
391
  # 区間推定
397
392
  #
398
- # @overload interval_estim(xi, a)
399
- # @param [array] xi データ(double[][][])
393
+ # @overload interval_estim(yi, xi, a)
394
+ # @param [array] yi データ(double[][])
395
+ # @param [array] xi データ(double[][])
400
396
  # @param [double] a 有意水準
401
397
  # @return [Hash] (min:下限信頼区間 max:上限信頼区間)
402
398
  # @example
399
+ # yi = [
400
+ # [3, 5, 3],
401
+ # [3, 3, 8],
402
+ # [2, 2, 2],
403
+ # [3, 4, 2],
404
+ # [1, 2, 0],
405
+ # ]
403
406
  # xi = [
404
- # [
405
- # [3,35], [5,38], [3,39],
406
- # ],
407
- # [
408
- # [3,36], [3,39], [8,54],
409
- # ],
410
- # [
411
- # [2,40], [2,45], [2,39],
412
- # ],
413
- # [
414
- # [3,47], [4,52], [2,48],
415
- # ],
416
- # [
417
- # [1,64], [2,80], [0,70],
418
- # ],
407
+ # [35, 38, 39],
408
+ # [36, 39, 54],
409
+ # [40, 45, 39],
410
+ # [47, 52, 48],
411
+ # [64, 80, 70],
419
412
  # ]
420
413
  # ancova = Num4AnovaLib::Num4AncovaLib.new
421
- # ancova.interval_estim(xi, 0.05)
414
+ # ancova.interval_estim(yi, xi, 0.05)
422
415
  # =>
423
416
  # {:min=>4.466605469341916, :max=>7.1909253948556096}
424
417
  # {:min=>5.05699825110459, :max=>6.386335082228742}
425
418
  # {:min=>2.510804295684195, :max=>4.250430272217034}
426
419
  # {:min=>2.8089257316042135, :max=>2.9566298239513418}
427
420
  # {:min=>-6.303283147572267, :max=>-0.6577045067487104}
428
- def interval_estim(xi, a)
421
+ def interval_estim(yi, xi, a)
429
422
  retRb = []
430
- retJava = @ancova.intervalEstim(xi.to_java(Java::double[][]), a)
423
+ retJava = @ancova.intervalEstim(yi.to_java(Java::double[]), xi.to_java(Java::double[]), a)
431
424
  sz = retJava.size
432
425
  sz.times do |i|
433
426
  retRb.push(
metadata CHANGED
@@ -1,14 +1,14 @@
1
1
  --- !ruby/object:Gem::Specification
2
2
  name: num4anova
3
3
  version: !ruby/object:Gem::Version
4
- version: 0.1.2
4
+ version: 0.2.1
5
5
  platform: java
6
6
  authors:
7
7
  - siranovel
8
8
  autorequire:
9
9
  bindir: bin
10
10
  cert_chain: []
11
- date: 2024-05-27 00:00:00.000000000 Z
11
+ date: 2024-06-17 00:00:00.000000000 Z
12
12
  dependencies:
13
13
  - !ruby/object:Gem::Dependency
14
14
  name: rake