num4regana 0.0.4-java → 0.0.6-java

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: d2c06b0a7caeac79ec2e1d451dccabd387449af22574e60520125374bd4296d7
-  data.tar.gz: 5744160fa6bec668de85a0ae0fd4b71625b6f4976a9b01e831ca71b4d654ec8c
+  metadata.gz: 07475023e86feba9a9a2013d572cdc433753da1cb570aa60fb3eb3662b09f3eb
+  data.tar.gz: 24a67fb47fe74071f39ae3d9d078071c1de5fce5b15afb74625354b34777b6be
 SHA512:
-  metadata.gz: 7030429dc48f2807211cf07446cd6fbcee4094fe55410d3d54ef01c9151ead39123ef869da446e6497db1f7e95ea59810281111a0f2a05922cdd3396fc9da6ba
-  data.tar.gz: 3967a10a750896e1b1a8f6bba7de42ae6a88862db18af456ba49f0018fc9e9e9d931aa792aceaa3e37d1c281dabfb52e461de484b0f349e59570e8e8ae73f3fe
+  metadata.gz: '058141be9257a86bc72b745402956142d015041e8812886a83ce2c49ac64993bc9fb1c864e2c7a093d0e42ba69f4dc45305f1eabb4fcffc4c0bc32d73333afb8'
+  data.tar.gz: db0acabd519684c4db318051df5467f50b0879aabdff44c426ff2c6e96559f762d286d1f8f7642052ff2dffa75847dc6f315610cad6c1d8fe268e14a251e6d22

data/CHANGELOG.md

@@ -2,6 +2,17 @@
 
 ## Unreleased
 
+## [0.0.6] - 2024-10-04
+
+### chg
+- chg  Hierarchical Bayesian models at EM algorithm
+
+## [0.0.5] - 2024-09-27
+
+### add
+- add Bayesian models
+- add Hierarchical Bayesian models
+
 ## [0.0.4] - 2024-09-13
 
 ### add

data/ext/num4regana/{AbstratGLM.java → AbstractGLM.java}

@@ -1,6 +1,6 @@
 import java.util.Arrays;
 
-abstract class AbstratGLM {
+abstract class AbstractGLM {
     private final double eta = 0.005;
     abstract double regression(double[] b, double[] xi);
     abstract double linkFunc(double q);
@@ -18,23 +18,21 @@ abstract class AbstratGLM {
     // AIC
     protected double calcAIC(double[] b, double[][] xij) {
         // 尤度計算
-        double maxL = calcL(b,xij); 
+        double maxL = calcLogL(b,xij); 
         int k = 1 + xij[0].length;
 
         return -2 * (maxL - k);
     }
     // 交差エントロピー計算
     private double[] calcE(double[] yi, double[] b, double[][] xij) {
-        double[] xi = new double[1 + xij[0].length];
-        double[] ei = new double[1 + xij[0].length];
+        double[] xi = new double[b.length];
+        double[] ei = new double[b.length];
 
         Arrays.fill(ei, 0.0);
         for(int i = 0; i < yi.length; i++) {
             xi[0] = 1.0;
             System.arraycopy(xij[i], 0, xi, 1, xij[0].length);
-
-            double q = regression(b, xi);
-            double p = linkFunc(q);
+            double p = linkFunc(regression(b, xi));
 
             for(int j = 0; j < xi.length; j++) {                
                 ei[j] += (p - yi[i]) * xi[j];
@@ -43,20 +41,20 @@ abstract class AbstratGLM {
 
         return ei;
     }
-    // 尤度計算(パラメータ)
-    private double calcL(double[] b, double[][] xij) {
+    // 対数尤度計算(パラメータ)
+    private double calcLogL(double[] b, double[][] xij) {
         double l = 0.0;
-        double[] xi = new double[1 + xij[0].length];
+        double[] xi = new double[b.length];
 
         for(int i = 0; i < xij.length; i++) {
             xi[0] = 1.0;
             System.arraycopy(xij[i], 0, xi, 1, xij[0].length);
-            double q = regression(b, xi);
-            double p = linkFunc(q);
+
+            double p = linkFunc(regression(b, xi));
 
             l += Math.log(p);
         }
         return l;
-    }   
+    }
 }
 

data/ext/num4regana/AbstractGLMM.java

@@ -0,0 +1,200 @@
+import java.util.Arrays;
+import org.apache.commons.math3.distribution.BetaDistribution;
+import org.apache.commons.math3.distribution.NormalDistribution;
+
+abstract class AbstractGLMM {
+    private NormalDistribution nDist = new NormalDistribution(0,1);
+    abstract double regression(double[] b, double[] xi, double r);
+    abstract double linkFunc(double q);
+    // mcmc法
+    // （メトロポリス法,ギブスサンプリング）
+    protected double[] mcmcGS(double[] yi, double[] b, double[][] xij) {
+        BetaDistribution beDist = new BetaDistribution(50, 50);
+        double[] newB = new double[b.length];
+
+        for(int i = 0; i < b.length; i++) {
+            newB = Arrays.copyOf(b, b.length);
+            newB[i] = beDist.sample();
+
+            b[i] = mcmcSample(
+                calcLx(b,xij),         // oldL
+                calcLx(newB,xij),      // newL
+                new double[] {b[i], newB[i]} // bTbl: [0]=> oldB, [1]=> newB
+            );
+        }
+        return b;
+    }
+    // BIC
+    protected double calcBIC(double[] b, double[][] xij) {
+        // 尤度計算
+        double maxL = calcLogLx(b,xij);
+        int k = 1 + xij[0].length;
+        int n = xij.length;
+
+        return -2 * maxL + k * Math.log(n);
+    }
+    // EMアルゴリズム
+    protected double[] mcmcEM(double[] yi, double[] b, double[][] xij) {
+        double[] newB = new double[b.length];
+        double[][] bE = calcEStep(yi, b, xij);
+        double[] bM = calcMStep(yi, bE, xij);
+
+        for(int i = 0; i < newB.length; i++) {
+            newB[i] = bM[i];  
+        }
+        return newB;
+    }
+    /*********************************/
+    /* interface define              */
+    /*********************************/
+    /*********************************/
+    /* class define                  */
+    /*********************************/
+    private static class ArraysFillEx {
+        public static void fill(Object array, Object value) {
+            // 第一引数が配列か判定
+            Class<?> type = array.getClass();
+            if (!type.isArray()) {
+                throw new IllegalArgumentException("not array");
+            }
+
+            // クラスの型を判定
+            String arrayClassName = array.getClass().getSimpleName()
+                    .replace("[]", "")
+                    .toLowerCase();
+            String valueClassName = value.getClass().getSimpleName()
+                    .toLowerCase()
+                    .replace("character", "char")
+                    .replace("integer", "int");
+            if (!arrayClassName.equals(valueClassName)) {
+                throw new IllegalArgumentException("does not matc");
+            }
+
+            // 処理
+            if (type.getComponentType().isArray()) {
+                for(Object o: (Object[])array) {
+                    fill(o, value);
+                }
+            }
+            else if (array instanceof boolean[]) {
+                Arrays.fill((boolean[])array, (boolean)value);
+            }
+            else if (array instanceof char[]) {
+                Arrays.fill((char[])array, (char)value);
+            }
+            else if (array instanceof byte[]) {
+                Arrays.fill((byte[])array, (byte)value);
+            }
+            else if (array instanceof short[]) {
+                Arrays.fill((short[])array, (short)value);
+            }
+            else if (array instanceof int[]) {
+                Arrays.fill((int[])array, (int)value);
+            }
+            else if (array instanceof long[]) {
+                Arrays.fill((long[])array, (long)value);
+            }
+            else if (array instanceof float[]) {
+                Arrays.fill((float[])array, (float)value);
+            }
+            else if (array instanceof double[]) {
+                Arrays.fill((double[])array, (double)value);
+            }
+            else {
+                Arrays.fill((Object[])array, value);
+            }
+        }
+    }
+    /* ------------------------------------------------------------------ */
+    private double mcmcSample(double oldL, double newL, double[] bTbl) {
+        double r = newL / oldL;
+        BetaDistribution beDist2 = new BetaDistribution(1, 1);  // 確率用
+        double b;
+
+        b = bTbl[0];
+        if (r > 1.0) {
+            b = bTbl[1];
+        }
+        else {
+            double r2 = beDist2.sample();
+
+            if (r2 < (1.0 - r)) {
+                b = bTbl[1];
+            }
+        }            
+        return b;
+    }
+    // 尤度計算(パラメータ)
+    private double calcLx(double[] b, double[][] xij) {
+        double l = 1.0;
+        double[] xi = new double[b.length];
+
+        for(int i = 0; i < xij.length; i++) {
+            xi[0] = 1.0;
+            System.arraycopy(xij[i], 0, xi, 1, xij[0].length);
+            double q = linkFunc(
+                regression(b, xi, nDist.sample())
+            );
+
+            l *= q;
+        }
+        return l;
+    }   
+    // 対数尤度計算(パラメータ)
+    private double calcLogLx(double[] b, double[][] xij) {
+        double l = 0.0;
+        double[] xi = new double[b.length];
+
+        for(int i = 0; i < xij.length; i++) {
+            xi[0] = 1.0;
+            System.arraycopy(xij[i], 0, xi, 1, xij[0].length);
+            double q = linkFunc(
+                regression(b, xi, nDist.sample())
+            );
+
+            l += Math.log(q);
+        }
+        return l;
+    }
+    // E-Step
+    //  (Expetation:自己エントロピー)
+    private double[][] calcEStep(double[] yi, double[] b, double[][] xij) {
+        double[][] bh = new double[yi.length][b.length];
+        double[] xi = new double[b.length];
+
+        ArraysFillEx.fill(bh, 0.0);
+        for(int i = 0; i < yi.length; i++) {
+            xi[0] = 1.0;
+            System.arraycopy(xij[i], 0, xi, 1, xij[i].length);
+            double p = yi[i];
+            double q = linkFunc(regression(b, xi, nDist.sample()));
+
+            for(int j = 0; j < b.length; j++) {
+                bh[i][j] = 
+                    Math.log(p * xi[j]) - q * (Math.log(q) - Math.log(p * xi[j]));
+            }
+        }
+        return bh;
+    }
+    // M-Step
+    //  (Maximiation:KLダイバージェンス)
+    private double[] calcMStep(double[] yi, double[][] q, double[][] xij) {
+        double[] xi = new double[1 + xij[0].length];
+        double[] ei = new double[1 + xij[0].length];
+
+        Arrays.fill(ei, 0.0);
+        for(int j = 0; j < xi.length; j++) {
+            for(int i = 0; i < xij.length; i++) {
+                xi[0] = 1.0;
+                System.arraycopy(xij[i], 0, xi, 1, xij[0].length);
+
+                double p = yi[i];
+
+                ei[j] += q[i][j] * (Math.log(p * xi[j]) - Math.log(q[i][j]));
+            }
+            ei[j] = -1 * ei[j];
+        }
+        return ei;
+    }   
+}
+

data/ext/num4regana/LogitBayesRegAna.java

@@ -0,0 +1,86 @@
+import java.util.Map;
+import java.util.Arrays;
+import org.apache.commons.math3.distribution.BetaDistribution;
+
+public class LogitBayesRegAna extends AbstractGLMM {
+    private final int NUM = 1000;
+    private final int TIM = 3;
+    private static LogitBayesRegAna regana = new LogitBayesRegAna();
+    public static LogitBayesRegAna getInstance() {
+        return regana;
+    }
+    public LineReg nonLineRegAna(double[] yi, double xij[][]) {
+        double[] b = initB(xij[0].length);
+
+        for  (int i = 0; i < NUM; i++) {
+            b = mcmcGS(yi, b, xij);
+        }
+
+        return new LineReg(b);
+    }
+    public double getBIC(Map<String, Object> regCoe, double[][] xij) {
+        double[] b = new double[1 + xij[0].length];
+
+        b[0] = (double)regCoe.get("intercept");
+        System.arraycopy(regCoe.get("slope"), 0, b, 1, xij[0].length);
+        return calcBIC(b, xij);
+    }
+    private double[] initB(int xsie) {
+        double[] b = new double[1 + xsie];
+        BetaDistribution beDist = new BetaDistribution(50, 50);
+
+        for(int i = 0; i < b.length; i++) {
+            b[i] = beDist.sample();
+        }
+        return b;
+    }
+    private double[] calcMeanBy(double[] yi, double[] b) {
+        double[] meanB = new double[b.length];
+
+        Arrays.fill(meanB, 0.0);
+        for(int i = 0; i < meanB.length; i++) {
+            for(int j = 0; j < yi.length; j++) {
+                meanB[i] += yi[j] * b[i];
+            }
+        }
+        return meanB;
+    }
+    // q = b0 + b1 * x0 + r
+    // (ランダム切片モデル)
+    double regression(double[] b, double[] xi, double r) {
+        double ret = 0.0;
+
+        for(int i = 0; i < xi.length; i++) {
+            ret += b[i] * xi[i];
+        }
+        return ret + r;
+    }
+    // p = 1 / (1 + exp( -q))
+    double linkFunc(double q) {
+        return 1.0 / (1.0 + Math.exp(-1.0 * q));
+    }
+    /*********************************/
+    /* interface define              */
+    /*********************************/
+    /*********************************/
+    /* class define                  */
+    /*********************************/
+    public class LineReg {
+        private double a   = 0.0;
+        private double[] b = null;
+        public LineReg(double[] b) {
+            this.a = b[0];
+            this.b = new double[b.length - 1];
+            for (int i = 0; i < this.b.length; i++) {
+                this.b[i] = b[i + 1];
+            }
+        }
+        public double getIntercept() {
+            return a;
+        }
+        public double[] getSlope() {
+            return b;
+        }
+    }
+}
+

data/ext/num4regana/LogitRegAna.java

@@ -1,7 +1,7 @@
 import java.util.Arrays;
 import java.util.Map;
 
-public class LogitRegAna extends AbstratGLM {
+public class LogitRegAna extends AbstractGLM {
     private final int NUM = 1000;
     private static LogitRegAna regana = new LogitRegAna();
     public static LogitRegAna getInstance() {

data/ext/num4regana/PoissonBayesRegAna.java

@@ -0,0 +1,75 @@
+import java.util.Arrays;
+import java.util.Map;
+import org.apache.commons.math3.distribution.BetaDistribution;
+
+public class PoissonBayesRegAna extends AbstractGLMM {
+    private final int NUM = 1000;
+    private final int TIM = 3;
+    private static PoissonBayesRegAna regana = new PoissonBayesRegAna();
+    public static PoissonBayesRegAna getInstance() {
+        return regana;
+    }
+    public LineReg nonLineRegAna(double[] yi, double xij[][]) {
+        double[] b = initB(xij[0].length);
+
+        for  (int i = 0; i < NUM; i++) {
+            b = mcmcGS(yi, b, xij);
+        }
+        return new LineReg(b);
+    }
+    public double getBIC(Map<String, Object> regCoe, double[][] xij) {
+        double[] b = new double[1 + xij[0].length];
+
+        b[0] = (double)regCoe.get("intercept");
+        System.arraycopy(regCoe.get("slope"), 0, b, 1, xij[0].length);
+        return calcBIC(b, xij);
+    }
+    private double[] initB(int xsie) {
+        double[] b = new double[1 + xsie];
+        BetaDistribution beDist = new BetaDistribution(50, 50);
+
+        for(int i = 0; i < b.length; i++) {
+            b[i] = beDist.sample();
+        }
+        return b;
+    }
+    // q = b0 + b1 * x0 + r
+    // (ランダム切片モデル)
+    double regression(double[] b, double[] xi, double r) {
+        double ret = 0.0;
+
+        for(int i = 0; i < xi.length; i++) {
+            ret += b[i] * xi[i];
+        }
+        return ret + r;
+    }
+    // p = exp(q)
+    double linkFunc(double q) {
+        return Math.exp(q);
+    }
+    /*********************************/
+    /* interface define              */
+    /*********************************/
+    /*********************************/
+    /* class define                  */
+    /*********************************/
+    public class LineReg {
+        private double a   = 0.0;
+        private double[] b = null;
+        public LineReg(double[] b) {
+            this.a = b[0];
+            this.b = new double[b.length - 1];
+            for (int i = 0; i < this.b.length; i++) {
+                this.b[i] = b[i + 1];
+            }
+        }
+        public double getIntercept() {
+            return a;
+        }
+        public double[] getSlope() {
+            return b;
+        }
+    }
+    
+}
+

data/ext/num4regana/PoissonHierBayesRegAna.java

@@ -0,0 +1,61 @@
+import java.util.Arrays;
+import java.util.Map;
+import org.apache.commons.math3.distribution.BetaDistribution;
+
+public class PoissonHierBayesRegAna extends AbstractGLMM {
+    private final int NUM = 1000;
+    private static PoissonHierBayesRegAna regana = new PoissonHierBayesRegAna();
+    public static PoissonHierBayesRegAna getInstance() {
+        return regana;
+    }
+    public LineReg nonLineRegAna(double[] yi, double xij[][]) {
+        double[] b = initB(xij[0].length);
+
+        for  (int i = 0; i < NUM; i++) {
+            b = mcmcEM(yi, b, xij);
+        }
+        return new LineReg(b);
+    }
+    private double[] initB(int xsie) {
+        double[] b = new double[1 + xsie];
+
+        Arrays.fill(b, 0.0);
+        return b;
+    }
+    // q = b0 + b1 * x0
+    double regression(double[] b, double[] xi, double r) {
+        double ret = 0.0;
+
+        for(int i = 0; i < xi.length; i++) {
+            ret += b[i] * xi[i];
+        }
+        return ret;
+    }
+    // p = exp(q)
+    double linkFunc(double q) {
+        return Math.exp(q);
+    }
+    /*********************************/
+    /* interface define              */
+    /*********************************/
+    /*********************************/
+    /* class define                  */
+    /*********************************/
+    public class LineReg {
+        private double a   = 0.0;
+        private double[] b = null;
+        public LineReg(double[] b) {
+            this.a = b[0];
+            this.b = new double[b.length - 1];
+            for (int i = 0; i < this.b.length; i++) {
+                this.b[i] = b[i + 1];
+            }
+        }
+        public double getIntercept() {
+            return a;
+        }
+        public double[] getSlope() {
+            return b;
+        }
+    }
+}

data/ext/num4regana/PoissonRegAna.java

@@ -1,7 +1,7 @@
 import java.util.Arrays;
 import java.util.Map;
 
-public class PoissonRegAna extends AbstratGLM {
+public class PoissonRegAna extends AbstractGLM {
     private final int NUM = 1000;
     private static PoissonRegAna regana = new PoissonRegAna();
     public static PoissonRegAna getInstance() {

data/ext/num4regana/ProBitRegAna.java

@@ -2,7 +2,7 @@ import java.util.Arrays;
 import org.apache.commons.math3.distribution.NormalDistribution;
 import java.util.Map;
 
-public class ProBitRegAna extends AbstratGLM {
+public class ProBitRegAna extends AbstractGLM {
     private final int NUM = 1000;
     private static ProBitRegAna regana = new ProBitRegAna();
     private NormalDistribution ndist = new NormalDistribution(0, 1);

data/lib/num4glmmregana.rb

@@ -0,0 +1,180 @@
+require 'java'
+require 'num4regana.jar'
+require 'commons-math3-3.6.1.jar'
+
+java_import 'LogitBayesRegAna'
+java_import 'PoissonBayesRegAna'
+java_import 'java.util.HashMap'
+
+# 一般化線形混合モデル
+#  (Apache commoms math3使用)
+module Num4GLMMRegAnaLib
+    # (2項)ベイズロジスティック回帰分析
+    class LogitBayesRegAnaLib
+        def initialize
+            @multana = LogitBayesRegAna.getInstance()
+        end
+        # (2項)ベイズロジスティック回帰分析
+        #
+        # @overload non_line_reg_ana(yi, xij)
+        #   @param [Array] yi yの値(double[])
+        #   @param [Array] xij xの値(double[][])
+        #   @return [Hash] (intercept:定数項 slope:回帰係数)
+        # @example
+        #   glsyi = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
+        #   glsxij = [
+        #        [1, 24],
+        #        [1, 18],
+        #        [0, 15],
+        #        [1, 16],
+        #        [0, 10],
+        #        [1, 26],
+        #        [1, 2],
+        #        [0, 24],
+        #        [1, 18],
+        #        [1, 22],
+        #        [1, 3],
+        #        [1, 6],
+        #        [0, 15],
+        #        [0, 12],
+        #        [1, 6],
+        #        [0, 6],
+        #        [1, 12],
+        #        [0, 12],
+        #        [1, 18],
+        #        [1, 3],
+        #        [1, 8],
+        #        [0, 9],
+        #        [0, 12],
+        #        [0, 6],
+        #        [0, 8],
+        #        [1, 12],
+        #   ]
+        #   regana = Num4GLMMRegAnaLib::LogitBayesRegAnaLib.new
+        #   regana.non_line_reg_ana(glsyi, glsxij)
+        #   => 
+        #     {  
+        #        :intercept=>0.5742886218005325,      # 定数項
+        #                                             # 回帰係数
+        #        :slope=>[0.5517212822536828, 0.5748054561700319]
+        #     }
+        def non_line_reg_ana(yi, xij)
+            multRet = @multana.nonLineRegAna(yi.to_java(Java::double), xij.to_java(Java::double[]))
+            retRb = {
+                "intercept":  multRet.getIntercept(), # 定数項
+                "slope":      multRet.getSlope().to_a,     # 回帰係数
+            }
+            return retRb
+        end
+        # BIC
+        #
+        # @overload get_bic(regcoe, xij)
+        #   @param [Hash] regcoe 回帰係数(intercept:定数項 slope:回帰係数)
+        #   @param [Array] xij xの値(double[][])
+        #   @return double BIC値
+        # @example
+        #   reg = {
+        #        :intercept=>  -6.2313,    # 定数項
+        #        :slope=>      [2.5995, 0.1652],     # 回帰係数
+        #   }
+        #   xij = [
+        #        [1, 24],
+        #        [1, 18],
+        #        [0, 15],
+        #        [1, 16],
+        #        [0, 10],
+        #        [1, 26],
+        #        [1, 2],
+        #        [0, 24],
+        #        [1, 18],
+        #        [1, 22],
+        #        [1, 3],
+        #        [1, 6],
+        #        [0, 15],
+        #        [0, 12],
+        #        [1, 6],
+        #        [0, 6],
+        #        [1, 12],
+        #        [0, 12],
+        #        [1, 18],
+        #        [1, 3],
+        #        [1, 8],
+        #        [0, 9],
+        #        [0, 12],
+        #        [0, 6],
+        #        [0, 8],
+        #        [1, 12],
+        #   ]
+        #   regana = Num4GLMMRegAnaLib::LogitBayesRegAnaLib.new
+        #   regana.get_bic(reg, xij)
+        #   => 159.386
+        def get_bic(regcoe, xij)
+            o = HashMap.new
+            o["intercept"] = regcoe[:intercept]
+            o["slope"]     = regcoe[:slope].to_java(Java::double)
+            @multana.getBIC(o, xij.to_java(Java::double[]))
+        end
+    end
+    # ベイズポアソン回帰分析
+    class PoissonBayesRegAnaLib
+        def initialize
+            @multana = PoissonBayesRegAna.getInstance()
+        end
+        # ベイズポアソン回帰分析
+        #
+        # @overload non_line_reg_ana(yi, xij)
+        #   @param [Array] yi yの値(double[])
+        #   @param [Array] xij xの値(double[][])
+        #   @return [Hash] (intercept:定数項 slope:回帰係数)
+        # @example
+        #   glsyi = [4, 10, 7, 14]
+        #   glsxij = [
+        #       [1],
+        #       [2],
+        #       [3],
+        #       [4],
+        #   ]
+        #   regana = Num4GLMMRegAnaLib::PoissonBayesRegAnaLib.new
+        #   regana.non_line_reg_ana(glsyi, glsxij)
+        #   => 
+        #     {
+        #        :intercept=>0.4341885635221602, # 定数項
+        #        :slope=>[0.5703137378188881]    # 回帰係数
+        #     }
+        def non_line_reg_ana(yi, xij)
+            multRet = @multana.nonLineRegAna(yi.to_java(Java::double), xij.to_java(Java::double[]))
+            retRb = {
+                "intercept":  multRet.getIntercept(), # 定数項
+                "slope":      multRet.getSlope().to_a,     # 回帰係数
+            }
+            return retRb
+        end
+        # BIC
+        #
+        # @overload get_bic(regcoe, xij)
+        #   @param [Hash] regcoe 回帰係数(intercept:定数項 slope:回帰係数)
+        #   @param [Array] xij xの値(double[][])
+        #   @return double BIC値
+        # @example
+        #   reg = {
+        #        :intercept=>0.4341885635221602, # 定数項
+        #        :slope=>[0.5703137378188881]    # 回帰係数
+        #   }
+        #   xij = [
+        #       [1],
+        #       [2],
+        #       [3],
+        #       [4],
+        #   ]
+        #   regana = Num4GLMMRegAnaLib::BayesPoissonRegAnaLib.new
+        #   regana.get_bic(reg, xij)
+        #   => -13.157
+        def get_bic(regcoe, xij)
+            o = HashMap.new
+            o["intercept"] = regcoe[:intercept]
+            o["slope"]     = regcoe[:slope].to_java(Java::double)
+            @multana.getBIC(o, xij.to_java(Java::double[]))
+        end
+    end
+end
+

data/lib/num4hbmregana.rb

@@ -0,0 +1,45 @@
+require 'java'
+require 'num4regana.jar'
+require 'commons-math3-3.6.1.jar'
+
+java_import 'PoissonHierBayesRegAna'
+# 階層ベイズモデル
+#  (Apache commoms math3使用)
+module Num4HBMRegAnaLib
+    # 階層ベイズポアソン回帰分析
+    class PoissonHierBayesRegAnaLib
+        def initialize
+            @multana = PoissonHierBayesRegAna.getInstance()
+        end
+        # ポアソン回帰分析
+        #
+        # @overload non_line_reg_ana(yi, xij)
+        #   @param [Array] yi yの値(double[])
+        #   @param [Array] xij xの値(double[][])
+        #   @return [Hash] (intercept:定数項 slope:回帰係数)
+        # @example
+        #   glsyi = [4, 10, 7, 14]
+        #   glsxij = [
+        #       [1],
+        #       [2],
+        #       [3],
+        #       [4],
+        #   ]
+        #   regana = Num4RegAnaLib::HierBayesPoissonRegAnaLib.new
+        #   regana.non_line_reg_ana(glsyi, glsxij)
+        #   => 
+        #     {
+        #        "intercept":  0.477366,    # 定数項
+        #        "slope":      [0.538545],  # 回帰係数
+        #     }
+        def non_line_reg_ana(yi, xij)
+            multRet = @multana.nonLineRegAna(yi.to_java(Java::double), xij.to_java(Java::double[]))
+            retRb = {
+                "intercept":  multRet.getIntercept(), # 定数項
+                "slope":      multRet.getSlope().to_a,     # 回帰係数
+            }
+            return retRb
+        end
+    end
+end
+

data/lib/num4regana.rb

@@ -1,10 +1,14 @@
 require_relative('num4lineregana')
 require_relative('num4glmregana')
+require_relative('num4glmmregana')
+require_relative('num4hbmregana')
 
 # 回帰分析
 module Num4RegAnaLib
     include Num4LineRegAnaLib
     include Num4GLMRegAnaLib
+    include Num4GLMMRegAnaLib
+    include Num4HBMRegAnaLib
 end
 
 

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: num4regana
 version: !ruby/object:Gem::Version
-  version: 0.0.4
+  version: 0.0.6
 platform: java
 authors:
 - siranovel
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2024-09-13 00:00:00.000000000 Z
+date: 2024-10-04 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake
@@ -61,15 +61,20 @@ files:
 - Gemfile
 - LICENSE
 - Rakefile
-- ext/num4regana/AbstratGLM.java
-- ext/num4regana/AbstratGLMM.java
+- ext/num4regana/AbstractGLM.java
+- ext/num4regana/AbstractGLMM.java
+- ext/num4regana/LogitBayesRegAna.java
 - ext/num4regana/LogitRegAna.java
 - ext/num4regana/MultRegAna.java
+- ext/num4regana/PoissonBayesRegAna.java
+- ext/num4regana/PoissonHierBayesRegAna.java
 - ext/num4regana/PoissonRegAna.java
 - ext/num4regana/ProBitRegAna.java
 - ext/num4regana/SmplRegAna.java
 - lib/commons-math3-3.6.1.jar
+- lib/num4glmmregana.rb
 - lib/num4glmregana.rb
+- lib/num4hbmregana.rb
 - lib/num4lineregana.rb
 - lib/num4regana.rb
 homepage: http://github.com/siranovel/num4regana

data/ext/num4regana/AbstratGLMM.java

@@ -1,18 +0,0 @@
-import java.util.Arrays;
-import org.apache.commons.math3.distribution.BetaDistribution;
-
-abstract class AbstratGLMM {
-    abstract double rereion(double[] b, double[] xi, double r);
-    abstract double linkFunc(double q);
-    protected double[] mcmc(double[] yi, double[] b, double[][] xij) {
-        double[] bnew = new double[1 + xij[0].length];
-        BetaDistribution beDist = new BetaDistribution(1,1);
-
-        for(int i= 0; i < bnew.length; i++) {
-System.out.printf("%f ", beDist.sample());
-        }
-        System.out.println();
-        return null;
-    }    
-}
-