num4tststatistic 0.0.4-java → 0.1.1-java

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 8b8a5b476ae860590efbcc631f0851e19d12b37f7f43e127d0637d7754912d4c
-  data.tar.gz: 2cd3007bd6872e547a7bc594d4d9d9af1b4b1ebcd8a45c12c69c5568752ff923
+  metadata.gz: 8035ed8852a010c953dc1f89787acb056eb516252687ed2a52ff558fb588ebf6
+  data.tar.gz: 61ed13288b017bf0845b3cc3bf6807694eb655666a71126a4b31c9b67fe61aa4
 SHA512:
-  metadata.gz: 1cf442842fe6bd7fd8447c12910c214aabdbe63aa0ab3811e984e289e5fd14276d99391672898c14c4a61368d403165f7a5e9d3c3cfccc9b640f095cb1f2f209
-  data.tar.gz: 7af0e381b6833ae5215d1048e7bdcb4fc8080d3fd0013eb3c7aabe868b5e44ebbb9ba5d89e5030a2428dd39d62e3f7d44f35772a162c4a6f3543c95aa9a6d25f
+  metadata.gz: 6031d267ff6f19ad8c12bc397cde2a6ae8aa95b0280564039c7f5a0c5cc8485455663bb7fd168d9358d8a180c17776440e620190c80af322ff8c21e863f622bc
+  data.tar.gz: f2214a6320e064958b12e7afd72162edad09a8f355690215028126bd9f230ed52dff4e048ca064393b115375c9499cbb7af59ed788dce0b33fbd41c82070eca5

data/CHANGELOG.md

@@ -2,6 +2,20 @@
 
 ## Unreleased
 
+## [0.1.1] - 2023-12-27
+
+### add
+- add function of utest.
+- add function of spearmanscorr.
+- add function of kendallscorr.
+- add function of ks2test.
+
+### change
+- chg function name from unCorrelation to pearsoCorrelation.
+
+### delete
+- del function of populationCorre.
+
 ## [0.0.4] - 2023-12-18
 
 ### add

data/ext/num4tststatistic/NonParametrixTest.java

@@ -0,0 +1,55 @@
+import java.util.Arrays;
+
+import org.apache.commons.math3.stat.inference.WilcoxonSignedRankTest;
+import org.apache.commons.math3.stat.inference.MannWhitneyUTest;
+import org.apache.commons.math3.stat.correlation.SpearmansCorrelation;
+import org.apache.commons.math3.stat.correlation.KendallsCorrelation;
+
+import org.apache.commons.math3.stat.inference.TestUtils;
+public class NonParametrixTest {
+    private static NonParametrixTest nonParamTest = new NonParametrixTest();
+    public static NonParametrixTest getInstance() {
+        return nonParamTest;
+    }
+    
+    // ウィルコクス符号付き順位検定
+    public double wilcoxon(double[] x, double[] y) {
+        int n = x.length;
+        WilcoxonSignedRankTest wilcox = new WilcoxonSignedRankTest();
+        double w = wilcox.wilcoxonSignedRank(x, y); 
+        double e_w = n * (n  + 1.0) / 4.0;
+        double var_w = n * (n + 1.0) * (2.0 * n + 1.0) / 24.0;
+
+        return (w - e_w) / Math.sqrt(var_w); 
+    }
+    // マン・ホイットニーのU検定
+    public double utest(double[] x, double[] y) {
+        int n1 = x.length;
+        int n2 = y.length;
+        MannWhitneyUTest utest = new MannWhitneyUTest();
+        double u = utest.mannWhitneyU(x, y);
+        double e_u = n1 * n2 / 2.0;
+        double var_u = (n1 * n2 * (n1 + n2 + 1.0)) / 12.0;
+
+        return (u - e_u) / Math.sqrt(var_u);
+    }
+    // スピアマンの順位相関係数
+    public double spearmanscorr(double[] x, double[] y) {
+        SpearmansCorrelation speamans = new SpearmansCorrelation();
+
+        return speamans.correlation(x, y);
+    }
+    // ケンドールの順位相関係数
+    public double kendallscorr(double[] x, double[] y) {
+        KendallsCorrelation kendalls = new KendallsCorrelation();
+
+        return kendalls.correlation(x, y);
+    }
+    // コルモゴルフ・スミルノフ検定(2標本)
+    public boolean ks2test(double[] xi1, double[] xi2, double a) {
+        double p = TestUtils.kolmogorovSmirnovTest(xi1, xi2);
+
+        return (p < a) ? true : false;
+    }
+}
+

data/ext/num4tststatistic/ParametrixTest.java

@@ -0,0 +1,193 @@
+import java.util.Arrays;
+import org.apache.commons.math3.stat.descriptive.SummaryStatistics;
+import org.apache.commons.math3.stat.correlation.PearsonsCorrelation;
+public class ParametrixTest {
+    private static ParametrixTest paramTest = new ParametrixTest();
+        
+    public static ParametrixTest getInstance() {
+        return paramTest;
+    }
+    // 正規母集団の母平均の検定量
+    public double populationMean(double[] xi, double m0) {
+        int n = xi.length;
+        SummaryStatistics stat = new SummaryStatistics();
+
+        Arrays.stream(xi).forEach(stat::addValue);
+        double m = stat.getMean();     // 平均
+        double s2 = stat.getVariance();// 分散
+        return (m - m0) / Math.sqrt(s2 / n);
+    }
+    // 正規母集団の母分散の検定量
+    public double populationVar(double[] xi, double sig0) {
+        int n = xi.length;
+        SummaryStatistics stat = new SummaryStatistics();
+        Arrays.stream(xi).forEach(stat::addValue);
+        double s2 = stat.getVariance();// 分散
+      
+        return (n - 1) * s2 / sig0;
+    }
+    // 母比率の検定量
+    public double populationRatio(int m, int n, double p0) {
+        double p = (double)m / (double)n;
+        return (p - p0) / Math.sqrt(p0 * (1-p0) / n);
+    }
+    // 2つの母平均の差の検定量
+    // (等分散性を仮定)
+    public double diffPopulationMean2EquVar(double[] xi1, double[] xi2) {
+        int n1 = xi1.length;
+        int n2 = xi2.length;
+        SummaryStatistics stat1 = new SummaryStatistics();
+        SummaryStatistics stat2 = new SummaryStatistics();
+        Arrays.stream(xi1).forEach(stat1::addValue);
+        Arrays.stream(xi2).forEach(stat2::addValue);
+
+        double m1 = stat1.getMean();
+        double m2 = stat2.getMean();
+        double s12 = stat1.getVariance();// 分散
+        double s22 = stat2.getVariance();// 分散
+        double s2 = ((n1 - 1) * s12 + (n2 - 1) * s22) / (n1 + n2 - 2);
+
+        return (m1 - m2) / Math.sqrt((1.0 / n1 + 1.0 / n2) * s2);
+    }
+    // 2つの母平均の差の検定量
+    // (不等分散性を仮定)
+    public double diffPopulationMean2UnEquVar(double[] xi1, double[] xi2) {
+        int n1 = xi1.length;
+        int n2 = xi2.length;
+        SummaryStatistics stat1 = new SummaryStatistics();
+        SummaryStatistics stat2 = new SummaryStatistics();
+        Arrays.stream(xi1).forEach(stat1::addValue);
+        Arrays.stream(xi2).forEach(stat2::addValue);
+
+        double m1 = stat1.getMean();
+        double m2 = stat2.getMean();
+        double s12 = stat1.getVariance();// 分散
+        double s22 = stat2.getVariance();// 分散
+
+        return (m1 - m2) / Math.sqrt(s12 / n1 + s22 / n2);
+    }
+    // ウェルチ検定の為の自由度
+    public int df4welch(double[] xi1, double[] xi2) {
+        int n1 = xi1.length;
+        int n2 = xi2.length;
+        SummaryStatistics stat1 = new SummaryStatistics();
+        SummaryStatistics stat2 = new SummaryStatistics();
+        Arrays.stream(xi1).forEach(stat1::addValue);
+        Arrays.stream(xi2).forEach(stat2::addValue);
+
+        double s12 = stat1.getVariance();// 分散
+        double s22 = stat2.getVariance();// 分散
+        double s14 = s12 * s12;
+        double s24 = s22 * s22;
+        int n12 = n1 * n1;
+        int n22 = n2 * n2;
+        double ns = (s12 / n1) + (s22 / n2);
+
+        return (int)
+        (
+           (ns * ns) / 
+           (
+             s14 / (n12 * (n1 - 1)) + s24 / (n22 * (n2 - 1))
+           )
+        );
+    }
+    // 対応のある2つの母平均の差の検定量
+    public double diffPopulationMean(double[] xi1, double[] xi2) {
+        int n = xi1.length;
+        SummaryStatistics stat = new SummaryStatistics();
+
+        for(int i = 0; i < n; i++) {
+            stat.addValue(xi1[i] - xi2[i]);
+        }
+        double m = stat.getMean();
+        double s2 = stat.getVariance();// 分散
+      
+        return (m - 0) / Math.sqrt(s2/n);
+    }
+    // 2つの母分散の差の検定量
+    public double diffPopulationVar(double[] xi1, double[] xi2) {
+        SummaryStatistics stat1 = new SummaryStatistics();
+        SummaryStatistics stat2 = new SummaryStatistics();
+        Arrays.stream(xi1).forEach(stat1::addValue);
+        Arrays.stream(xi2).forEach(stat2::addValue);
+
+        double s12 = stat1.getVariance();// 分散
+        double s22 = stat2.getVariance();// 分散
+        return s12 / s22;
+    }
+    // 2つの母比率の差の検定量
+    public double diffPopulationRatio(int m1, int n1, int m2, int n2) {
+        double p1 = (double)m1 / (double)n1;
+        double p2 = (double)m2 / (double)n2;
+        double p = (double)(m1 + m2) / (double)(n1 + n2);
+
+        return (p1 - p2) / Math.sqrt(p * (1 - p) * (1.0 / n1 + 1.0 / n2));
+    }
+    // ピアソン相関係数
+    public double pearsoCorrelation(double[] x, double[] y) {
+        PearsonsCorrelation corel = new PearsonsCorrelation();
+
+        return corel.correlation(x, y);
+    }
+    // 適合度の検定量
+    public double fidelity(double[] fi, double[] pi) {
+        double[] e = new double[fi.length];
+        double t = 0.0;
+        SummaryStatistics stat = new SummaryStatistics();
+
+        Arrays.stream(fi).forEach(stat::addValue);
+        double s = stat.getSum();
+
+        for (int i = 0; i < fi.length; i++) {
+            e[i] = s * pi[i];
+            double f_e = fi[i] - e[i];
+            t += f_e * f_e / e[i];
+        }
+        return t;
+    }
+    // 独立性の検定量
+    public double independency(double[][] fij) {
+        Independency indep = new Independency(fij);
+
+        indep.calsCells();
+        double t = indep.calcStatistic();
+
+        return t;
+    }
+    private class Independency {
+        private double[] fa = null;
+        private double[] fb = null;
+        private double[][] fij = null;
+        private long n = 0;
+        public Independency(double[][] fij) {
+            fa = new double[fij.length];
+            fb = new double[fij[0].length];
+            this.fij = fij;
+        }
+        // 各セルの計算
+        public void calsCells() {
+            for (int i = 0; i < fij.length; i++) {
+                fa[i] = 0.0;
+                for (int j = 0; j < fij[i].length; j++)  {
+                    fa[i] += fij[i][j];
+                    fb[j] += fij[i][j];
+                    n += fij[i][j];
+                }
+            }
+        }
+        // 検定統計量計算
+        public double calcStatistic() {
+            double t = 0.0;
+
+            for (int i = 0; i < fij.length; i++) {
+                for (int j = 0; j < fij[i].length; j++)  {
+                    double f_e = n * fij[i][j] - fa[i] * fb[j];
+
+                    t += f_e * f_e / (n * fa[i] * fb[j]);
+                }
+            }
+            return t;
+        }
+    }
+}
+

data/ext/num4tststatistic/TstStatistic.java

@@ -1,177 +1,7 @@
 import java.util.Arrays;
 
 import org.apache.commons.math3.stat.descriptive.SummaryStatistics;
-import org.apache.commons.math3.stat.correlation.PearsonsCorrelation;
-import org.apache.commons.math3.stat.inference.WilcoxonSignedRankTest;
 public class TstStatistic {
-    public static double populationMean(double[] xi, double m0) {
-        int n = xi.length;
-        SummaryStatistics stat = new SummaryStatistics();
-
-        Arrays.stream(xi).forEach(stat::addValue);
-        double m = stat.getMean();     // 平均
-        double s2 = stat.getVariance();// 分散
-        return (m - m0) / Math.sqrt(s2 / n);
-    }
-    public static double populationVar(double[] xi, double sig0) {
-        int n = xi.length;
-        SummaryStatistics stat = new SummaryStatistics();
-        Arrays.stream(xi).forEach(stat::addValue);
-        double s2 = stat.getVariance();// 分散
-        
-        return (n - 1) * s2 / sig0;
-    }
-    public static double populationRatio(int m, int n, double p0) {
-        double p = (double)m / (double)n;
-
-        return (p - p0) / Math.sqrt(p0 * (1-p0) / n);
-    }
-    public static double diffPopulationMean2EquVar(double[] xi1, double[] xi2) {
-        int n1 = xi1.length;
-        int n2 = xi2.length;
-        SummaryStatistics stat1 = new SummaryStatistics();
-        SummaryStatistics stat2 = new SummaryStatistics();
-        Arrays.stream(xi1).forEach(stat1::addValue);
-        Arrays.stream(xi2).forEach(stat2::addValue);
-
-        double m1 = stat1.getMean();
-        double m2 = stat2.getMean();
-        double s12 = stat1.getVariance();// 分散
-        double s22 = stat2.getVariance();// 分散
-        double s2 = ((n1 - 1) * s12 + (n2 - 1) * s22) / (n1 + n2 - 2);
-
-        return (m1 - m2) / Math.sqrt((1.0 / n1 + 1.0 / n2) * s2);
-    }
-    public static double diffPopulationMean2UnEquVar(double[] xi1, double[] xi2) {
-        int n1 = xi1.length;
-        int n2 = xi2.length;
-        SummaryStatistics stat1 = new SummaryStatistics();
-        SummaryStatistics stat2 = new SummaryStatistics();
-        Arrays.stream(xi1).forEach(stat1::addValue);
-        Arrays.stream(xi2).forEach(stat2::addValue);
-
-        double m1 = stat1.getMean();
-        double m2 = stat2.getMean();
-        double s12 = stat1.getVariance();// 分散
-        double s22 = stat2.getVariance();// 分散
-
-        return (m1 - m2) / Math.sqrt(s12 / n1 + s22 / n2);
-    }
-    public static int df4welch(double[] xi1, double[] xi2) {
-        int n1 = xi1.length;
-        int n2 = xi2.length;
-        SummaryStatistics stat1 = new SummaryStatistics();
-        SummaryStatistics stat2 = new SummaryStatistics();
-        Arrays.stream(xi1).forEach(stat1::addValue);
-        Arrays.stream(xi2).forEach(stat2::addValue);
-
-        double s12 = stat1.getVariance();// 分散
-        double s22 = stat2.getVariance();// 分散
-        double s14 = s12 * s12;
-        double s24 = s22 * s22;
-        int n12 = n1 * n1;
-        int n22 = n2 * n2;
-        double ns = (s12 / n1) + (s22 / n2);
-
-        return (int)
-        (
-           (ns * ns) / 
-           (
-             s14 / (n12 * (n1 - 1)) + s24 / (n22 * (n2 - 1))
-           )
-         );
-    }
-    public static double diffPopulationMean(double[] xi1, double[] xi2) {
-        int n = xi1.length;
-        SummaryStatistics stat = new SummaryStatistics();
-
-        for(int i = 0; i < n; i++) {
-            stat.addValue(xi1[i] - xi2[i]);
-        }
-        double m = stat.getMean();
-        double s2 = stat.getVariance();// 分散
-        
-        return (m - 0) / Math.sqrt(s2/n);
-    }
-    public static double diffPopulationVar(double[] xi1, double[] xi2) {
-        SummaryStatistics stat1 = new SummaryStatistics();
-        SummaryStatistics stat2 = new SummaryStatistics();
-        Arrays.stream(xi1).forEach(stat1::addValue);
-        Arrays.stream(xi2).forEach(stat2::addValue);
-
-        double s12 = stat1.getVariance();// 分散
-        double s22 = stat2.getVariance();// 分散
-        return s12 / s22;
-    }
-    public static double diffPopulationRatio(int m1, int n1, int m2, int n2) {
-        double p1 = (double)m1 / (double)n1;
-        double p2 = (double)m2 / (double)n2;
-        double p = (double)(m1 + m2) / (double)(n1 + n2);
-
-        return (p1 - p2) / Math.sqrt(p * (1 - p) * (1.0 / n1 + 1.0 / n2));
-    }
-    public static double unCorrelation(double[] x, double[] y) {
-        int n = x.length;
-        PearsonsCorrelation corel = new PearsonsCorrelation();
-
-        double r = corel.correlation(x, y);
-
-        return r * Math.sqrt(n - 2.0) / Math.sqrt(1.0 - r * r);
-    }
-    public static double populationCorre(double[] x, double[] y, double rth0) {
-        int n = x.length;
-        PearsonsCorrelation corel = new PearsonsCorrelation();
-
-        double r = corel.correlation(x, y);
-
-        return Math.sqrt(n-3.0) * 
-               (
-               0.5 * Math.log((1.0 + r) / (1.0 - r)) 
-             - 0.5 * Math.log((1.0 + rth0) / (1.0 - rth0))
-               );
-    }
-    public static double fidelity(double[] fi, double[] pi) {
-        double[] e = new double[fi.length];
-        double t = 0.0;
-        SummaryStatistics stat = new SummaryStatistics();
-
-        Arrays.stream(fi).forEach(stat::addValue);
-        double s = stat.getSum();
-
-        for (int i = 0; i < fi.length; i++) {
-            e[i] = s * pi[i];
-            double f_e = fi[i] - e[i];
-            t += f_e * f_e / e[i];
-        }
-        return t;
-    }
-    public static double independency(double[][] fij) {
-        double t = 0.0;
-        long n = 0;
-        double[] fa = new double[fij.length];
-        double[] fb = new double[fij[0].length];
-
-        // 各セルの計算
-        for (int i = 0; i < fij.length; i++) {
-            fa[i] = 0.0;
-            for (int j = 0; j < fij[i].length; j++)  {
-                fa[i] += fij[i][j];
-                fb[j] += fij[i][j];
-                n += fij[i][j];
-            }
-        }
-
-        // 検定統計量計算
-        for (int i = 0; i < fij.length; i++) {
-            for (int j = 0; j < fij[i].length; j++)  {
-                double f_e = n * fij[i][j] - fa[i] * fb[j];
-
-                t += f_e * f_e / (n * fa[i] * fb[j]);
-            }
-        }
-
-        return t;
-    }
     public static double grubbs(double[] xi, double xk) {
         SummaryStatistics stat = new SummaryStatistics();
 
@@ -186,11 +16,6 @@ public class TstStatistic {
         if (xk == max) { t = (xk - m) / sd; }
         return t;
     }
-    public static double wilcoxon(double[] x, double[] y) {
-        WilcoxonSignedRankTest wilcox = new WilcoxonSignedRankTest();
-
-        return wilcox.wilcoxonSignedRank(x, y); 
-    }
 }
 
 

data/lib/num4tststatistic.rb

@@ -3,11 +3,16 @@ require 'num4tststatistic.jar'
 require 'commons-math3-3.6.1.jar'
 
 java_import 'TstStatistic'
+java_import 'ParametrixTest'
+java_import 'NonParametrixTest'
 
 # 検定統計量を計算
 #  (Apache commoms math3使用)
 module Num4TstStatisticLib
-    class << self
+    class ParametrixTestLib
+        def initialize
+            @paramTest = ParametrixTest.getInstance()
+        end
         # 正規母集団の母平均の検定量
         #
         # @overload populationMean(xi, m0)
@@ -16,12 +21,13 @@ module Num4TstStatisticLib
         #   @return [double] 検定統計量
         # @example
         #   xi = [15.5, 15.7, 15.4, 15.4, 15.6, 15.4, 15.6, 15.5, 15.4]
-        #   Num4TstStatisticLib.populationMean(xi, 15.4)
+        #   paraTest = Num4TstStatisticLib::ParametrixTestLib.new
+        #   paraTest.populationMean(xi, 15.4)
         #   => 2.683
         # @note
         #   自由度(N-1)のt分布に従う
         def populationMean(xi, m0)
-            return TstStatistic.populationMean(xi.to_java(Java::double), m0)
+            return @paramTest.populationMean(xi.to_java(Java::double), m0)
         end
         # 正規母集団の母分散の検定量
         #
@@ -32,26 +38,29 @@ module Num4TstStatisticLib
         # @example
         #   xi = xi = [35.2, 34.5, 34.9, 35.2, 34.8, 35.1, 34.9, 35.2, 34.9, 34.8]
         #   sd = 0.4
-        #   Num4TstStatisticLib.populationVar(xi, sd*sd)
+        #   paraTest = Num4TstStatisticLib::ParametrixTestLib.new
+        #   paraTest.populationVar(xi, sd*sd)
         #   => 2.906
         # @note
         #   自由度(N-1)の階2乗分布に従う
         def populationVar(xi, sig0)
-            return TstStatistic.populationVar(xi.to_java(Java::double), sig0)
+            return @paramTest.populationVar(xi.to_java(Java::double), sig0)
         end
         # 母比率の検定量
+        #
         # @overload populationRatio(m, n, p0)
         #   @param [int]     m m値
         #   @param [int]     n N値
         #   @param [double]  p0 母比率
         #   @return [double] 検定統計量
         # @example
-        #   Num4TstStatisticLib.populationRatio(29, 346, 0.12)
+        #   paraTest = Num4TstStatisticLib::ParametrixTestLib.new
+        #   paraTest.populationRatio(29, 346, 0.12)
         #   => -2.071
         # @note
         #   標準正規分布 N(0,1*1)に従う
         def populationRatio(m, n, p0)
-            return TstStatistic.populationRatio(m, n, p0)
+            return @paramTest.populationRatio(m, n, p0)
         end
         # 2つの母平均の差の検定量
         # (等分散性を仮定)
@@ -63,12 +72,13 @@ module Num4TstStatisticLib
         # @example
         #   xi1 = [165, 130, 182, 178, 194, 206, 160, 122, 212, 165, 247, 195]
         #   xi2 = [180, 180, 235, 270, 240, 285, 164, 152]
-        #   Num4TstStatisticLib.diffPopulationMean2EquVar(xi1, xi2)
+        #   paraTest = Num4TstStatisticLib::ParametrixTestLib.new
+        #   paraTest.diffPopulationMean2EquVar(xi1, xi2)
         #   => -1.765
         # @note
         #   N1+N2-2のt分布に従う
         def diffPopulationMean2EquVar(xi1, xi2)
-            return TstStatistic.diffPopulationMean2EquVar(
+            return @paramTest.diffPopulationMean2EquVar(
                 xi1.to_java(Java::double), xi2.to_java(Java::double)
             )
         end
@@ -82,16 +92,18 @@ module Num4TstStatisticLib
         # @example
         #   xi1 = [165, 130, 182, 178, 194, 206, 160, 122, 212, 165, 247, 195]
         #   xi2 = [180, 180, 235, 270, 240, 285, 164, 152]
-        #   Num4TstStatisticLib.diffPopulationMean2UnEquVar(xi1, xi2)
+        #   paraTest = Num4TstStatisticLib::ParametrixTestLib.new
+        #   paraTest.diffPopulationMean2UnEquVar(xi1, xi2)
         #   => -1.636
         # @note
         #   df4welch関数で求めた自由度のt分布に従う
         def diffPopulationMean2UnEquVar(xi1, xi2)
-            return TstStatistic.diffPopulationMean2UnEquVar(
+            return @paramTest.diffPopulationMean2UnEquVar(
                 xi1.to_java(Java::double), xi2.to_java(Java::double)
             )
         end
         # ウェルチ検定の為の自由度
+        #
         # @overload df4welch(xi1, xi2)
         #   @param [Array] xi1 x1のデータ(double[])
         #   @param [Array] xi2 x2のデータ(double[])
@@ -99,10 +111,11 @@ module Num4TstStatisticLib
         # @example
         #   xi1 = [165, 130, 182, 178, 194, 206, 160, 122, 212, 165, 247, 195]
         #   xi2 = [180, 180, 235, 270, 240, 285, 164, 152]
-        #   Num4TstStatisticLib.df4welch(xi1, xi2)
+        #   paraTest = Num4TstStatisticLib::ParametrixTestLib.new
+        #   paraTest.df4welch(xi1, xi2)
         #   => 11
         def df4welch(xi1, xi2)
-            return TstStatistic.df4welch(
+            return @paramTest.df4welch(
                 xi1.to_java(Java::double), xi2.to_java(Java::double)
             )
         end
@@ -115,12 +128,13 @@ module Num4TstStatisticLib
         # @example
         #   xi1 = [37.1, 36.2, 36.6, 37.4, 36.8, 36.7, 36.9, 37.4, 36.6, 36.7]
         #   xi2 = [36.8, 36.6, 36.5, 37.0, 36.0, 36.5, 36.6, 37.1, 36.4, 36.7]
-        #   Num4TstStatisticLib.diffPopulationMean(xi1, xi2)
+        #   paraTest = Num4TstStatisticLib::ParametrixTestLib.new
+        #   paraTest.diffPopulationMean(xi1, xi2)
         #   => 2.283
         # @note
         #   自由度(N-1)のt分布に従う
-         def diffPopulationMean(xi1, xi2)
-            return TstStatistic.diffPopulationMean(
+        def diffPopulationMean(xi1, xi2)
+            return @paramTest.diffPopulationMean(
                 xi1.to_java(Java::double), xi2.to_java(Java::double)
             )
         end
@@ -133,12 +147,13 @@ module Num4TstStatisticLib
         # @example
         #   xi1 = [165, 130, 182, 178, 194, 206, 160, 122, 212, 165, 247, 195]
         #   xi2 = [180, 180, 235, 270, 240, 285, 164, 152]
-        #   Num4TstStatisticLib.diffPopulationVar(xi1, xi2)
+        #   paraTest = Num4TstStatisticLib::ParametrixTestLib.new
+        #   paraTest.diffPopulationVar(xi1, xi2)
         #   => 0.4727
         # @note
         #   自由度(N1-1,N2-1)のF分布に従う
         def diffPopulationVar(xi1, xi2)
-            return TstStatistic.diffPopulationVar(
+            return @paramTest.diffPopulationVar(
                 xi1.to_java(Java::double), xi2.to_java(Java::double)
             )
         end
@@ -151,50 +166,32 @@ module Num4TstStatisticLib
         #   @param [int] n2 N2値
         #   @return [double] 検定統計量
         # @example
-        #   Num4TstStatisticLib.diffPopulationRatio(469, 1200, 308, 900)
+        #   paraTest = Num4TstStatisticLib::ParametrixTestLib.new
+        #   paraTest.diffPopulationRatio(469, 1200, 308, 900)
         #   => 2.283
         # @note
         #   標準正規分布 N(0,1*1)に従う
         def diffPopulationRatio(m1, n1, m2, n2)
-            return TstStatistic.diffPopulationRatio(m1, n1, m2, n2)
+            return @paramTest.diffPopulationRatio(m1, n1, m2, n2)
         end
-        # 無相関の検定量
+        # ピアソン相関係数
+        #  (相関係数の検定)
         #
-        # @overload unCorrelation(x, y)
+        # @overload pearsoCorrelation(x, y)
         #   @param [Array] x xのデータ(double[])
         #   @param [Array] y yのデータ(double[])
-        #   @return [double] 検定統計量
+        #   @return [double] 相関係数
         # @example
         #   x = [113, 64, 16, 45, 28, 19, 30, 82, 76]
         #   y = [31, 5, 2, 17, 18, 2, 9, 25, 13]
-        #   Num4TstStatisticLib.unCorrelation(x, y)
+        #   paraTest = Num4TstStatisticLib::ParametrixTestLib.new
+        #   paraTest.pearsoCorrelation(x, y)
         #   => 3.1035
-        # @note
-        #   自由度(N-2)t分布に従う
-        def unCorrelation(x, y)
-            return TstStatistic.unCorrelation(
+        def pearsoCorrelation(x, y)
+            return @paramTest.pearsoCorrelation(
                 x.to_java(Java::double), y.to_java(Java::double)
             )
         end
-        # 母相関係数の検定量
-        #
-        # @overload populationCorre(x, y, rth0)
-        #   @param [Array] x xのデータ(double[])
-        #   @param [Array] y yのデータ(double[])
-        #   @param [double] rth0 母相関係数
-        #   @return [double] 検定統計量
-        # @example
-        #   x = [2750, 2956, 2675, 3198, 1816, 2233, 2375, 2288, 1932, 2036, 2183, 2882]
-        #   y = [249, 713, 1136, 575, 5654, 2107, 915, 4193, 7225, 3730, 472, 291]
-        #   Num4TstStatisticLib.populationCorre(x, y, -0.3)
-        #   => -2.107168
-        # @note
-        #   標準正規分布 N(0,1*1)に従う
-        def populationCorre(x, y, rth0)
-            return TstStatistic.populationCorre(
-                x.to_java(Java::double), y.to_java(Java::double), rth0
-            )
-        end
         # 適合度の検定量
         #
         # @overload fidelity(fi, pi)
@@ -204,12 +201,13 @@ module Num4TstStatisticLib
         # @example
         #   fi = [57, 33, 46, 14]
         #   pi = [0.4, 0.2, 0.3, 0.1]
-        #   Num4TstStatisticLib.fidelity(fi, pi)
+        #   paraTest = Num4TstStatisticLib::ParametrixTestLib.new
+        #   paraTest.fidelity(fi, pi)
         #   => 0.5389
         # @note
         #   自由度(n-1)の階２乗分布に従う
         def fidelity(fi, pi)
-            return TstStatistic.fidelity(fi.to_java(Java::double), pi.to_java(Java::double))
+            return @paramTest.fidelity(fi.to_java(Java::double), pi.to_java(Java::double))
         end
         # 独立性の検定量
         #
@@ -221,43 +219,121 @@ module Num4TstStatisticLib
         #     [57, 33, 46, 14],
         #     [89, 24, 75, 12],
         #   ]
-        #   Num4TstStatisticLib.independency(fij)
+        #   paraTest = Num4TstStatisticLib::ParametrixTestLib.new
+        #   paraTest.independency(fij)
         #   => 8.5711
         # @note
         #   自由度(m-1)(n-1)の階２乗分布に従う
         def independency(fij)
-            return TstStatistic.independency(fij.to_java(Java::double[]))
+            return @paramTest.independency(fij.to_java(Java::double[]))
         end
-        # グラプス・スミルノフの外れ値の検定量
+    end
+    class NonParametrixTestLib
+        def initialize
+            @nonParamTest = NonParametrixTest.getInstance()
+        end
+        # マン・ホイットニーのU検定
         #
-        # @overload grubbs(xi, xk)
-        #   @param [Array] xi xiのデータ(double[])
-        #   @param [double] xk 外れ値
+        # @overload utest(x, y)
+        #   @param [Array] x xのデータ(double[])
+        #   @param [Array] y yのデータ(double[])
         #   @return [double] 検定統計量
         # @example
-        #   xi = [3.4, 3.5, 3.3, 2.2, 3.3, 3.4, 3.6, 3.2]
-        #   Num4TstStatisticLib.grubbs(xi, 2.2)
-        #   => 2.3724
+        #   x = [165, 130, 182, 178, 194, 206, 160, 122, 212, 165, 247, 195]
+        #   y = [180, 180, 235, 270, 240, 285, 164, 152]
+        #   nonParaTest = Num4TstStatisticLib::NonParametrixTestLib.new
+        #   nonParaTest.utest(x, y)
+        #   => 63.0
         # @note
-        #   グラプス・スミルノフの数表に従う
-        def grubbs(xi, xk)
-            return TstStatistic.grubbs(xi.to_java(Java::double), xk)
+        #   近似的に標準正規分布 N(0,1*1)に従う
+        def utest(x, y)
+            return @nonParamTest.utest(x.to_java(Java::double), y.to_java(Java::double))
         end
         # ウィルコクス符号付き順位検定
         #
-        # @overload wilcoxon(x, y)
+        # @overload wilcoxontest(x, y)
         #   @param [Array] x xのデータ(double[])
         #   @param [Array] y yのデータ(double[])
         #   @return [double] 検定統計量
         # @example
-        #   x = [28, 25, 29, 28, 30, 20, 31, 27, 24, 26, 35, 23, 27, 32]
-        #   y = [32, 30, 31, 27, 35, 25, 40, 30, 45, 28, 32, 30, 30, 38]
-        #   Num4TstStatisticLib.wilcoxon(x, y)
-        #   => 99
+        #   x = [37.1, 36.2, 36.6, 37.4, 36.8, 36.7, 36.9, 37.4, 36.6, 36.7]
+        #   y = [36.8, 36.6, 36.5, 37.0, 36.0, 36.5, 36.6, 37.1, 36.4, 36.7]
+        #   nonParaTest = Num4TstStatisticLib::NonParametrixTestLib.new
+        #   nonParaTest.wilcoxon(x, y)
+        #   => 46.5
         # @note
-        #   標準正規分布 N(0,1*1)に従う
+        #   近似的に標準正規分布 N(0,1*1)に従う
         def wilcoxon(x, y)
-            return TstStatistic.wilcoxon(x.to_java(Java::double), y.to_java(Java::double))
+            return @nonParamTest.wilcoxon(x.to_java(Java::double), y.to_java(Java::double))
+        end
+        # スピアマンの順位相関係数
+        #
+        # @overload spearmanscorr(x, y)
+        #   @param [Array] x xのデータ(double[])
+        #   @param [Array] y yのデータ(double[])
+        #   @return [double] 相関係数
+        # @example
+        #   x = [113, 64, 16, 45, 28, 19, 30, 82, 76]
+        #   y = [31, 5, 2, 17, 18, 2, 9, 25, 13]
+        #   nonParaTest = Num4TstStatisticLib::NonParametrixTestLib.new
+        #   nonParaTest.spearmanscorr(x, y)
+        #   => 0.745
+        # @note
+        #   無相関検定
+        def spearmanscorr(x, y)
+            return @nonParamTest.spearmanscorr(x.to_java(Java::double), y.to_java(Java::double))
+        end
+        # ケンドールの順位相関係数
+        #
+        # @overload kendallscorr(x, y)
+        #   @param [Array] x xのデータ(double[])
+        #   @param [Array] y yのデータ(double[])
+        #   @return [double] 相関係数
+        # @example
+        #   x = [113, 64, 16, 45, 28, 19, 30, 82, 76]
+        #   y = [31, 5, 2, 17, 18, 2, 9, 25, 13]
+        #   nonParaTest = Num4TstStatisticLib::NonParametrixTestLib.new
+        #   nonParaTest.kendallscorr(x, y)
+        #   => 0.592
+        # @note
+        #   無相関検定
+        def kendallscorr(x, y)
+            return @nonParamTest.kendallscorr(x.to_java(Java::double), y.to_java(Java::double))
+        end
+        # コルモゴルフ・スミルノフ検定(2標本)
+        #
+        # @overload ks2test(xi1, xi2)
+        #   @param [Array] xi1 x1のデータ(double[])
+        #   @param [Array] xi2 x2のデータ(double[])
+        #   @param [double] xi2 x2のデータ(double[])
+        #   @return [boolean] 検定結果(true:棄却域内 false:棄却域外)
+        # @example
+        #   xi1 = [165, 130, 182, 178, 194, 206, 160, 122, 212, 165, 247, 195]
+        #   xi2 = [180, 180, 235, 270, 240, 285, 164, 152]
+        #   nonParaTest = Num4TstStatisticLib::NonParametrixTestLib.new
+        #   nonParaTest.ks2test(xi1, xi2, 0.05)
+        #   => false
+        # @note
+        #   N1+N2-2のt分布に従う
+        def ks2test(xi1, xi2, a)
+            return @nonParamTest.ks2test(xi1.to_java(Java::double), xi2.to_java(Java::double), a)
+        end
+    end
+    class << self
+        # グラプス・スミルノフの外れ値の検定量
+        #
+        # @overload grubbs(xi, xk)
+        #   @param [Array] xi xiのデータ(double[])
+        #   @param [double] xk 外れ値
+        #   @return [double] 検定統計量
+        # @example
+        #   xi = [3.4, 3.5, 3.3, 2.2, 3.3, 3.4, 3.6, 3.2]
+        #   Num4TstStatisticLib.grubbs(xi, 2.2)
+        #   => 2.3724
+        # @note
+        #   外れ値の検定に従う
+        def grubbs(xi, xk)
+            return TstStatistic.grubbs(xi.to_java(Java::double), xk)
         end
     end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: num4tststatistic
 version: !ruby/object:Gem::Version
-  version: 0.0.4
+  version: 0.1.1
 platform: java
 authors:
 - siranovel
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2023-12-18 00:00:00.000000000 Z
+date: 2023-12-27 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake
@@ -61,6 +61,8 @@ files:
 - Gemfile
 - LICENSE
 - Rakefile
+- ext/num4tststatistic/NonParametrixTest.java
+- ext/num4tststatistic/ParametrixTest.java
 - ext/num4tststatistic/TstStatistic.java
 - lib/commons-math3-3.6.1.jar
 - lib/num4tststatistic.rb