栏目分类:
子分类:
返回
名师互学网用户登录
快速导航关闭
当前搜索
当前分类
子分类
实用工具
热门搜索
名师互学网 > IT > 软件开发 > 后端开发 > Java

一元函数之最小二乘回归+Java代码实现

Java 更新时间: 发布时间: IT归档 最新发布 模块sitemap 名妆网 法律咨询 聚返吧 英语巴士网 伯小乐 网商动力

一元函数之最小二乘回归+Java代码实现

代码
public class LeastSquaresRegression {

    double[] xData;
    double[] yData;

    public static void main(String[] args) {
        double[] xData = new double[]{1, 2, 3, 4,5,6,7,8,9,10,11,12};
        double[] yData = new double[]{4200,4300,4000,4400,5000,4700,5300,4900,5400,5700,6300,6000};
        double[] preX = {13,14,15,16,17,18,19,20,21,22,23,24};
        new LeastSquaresRegression(xData, yData).predict(preX);
    }

    public void predict(double[] preX) {
        // 计算n
        int n = xData.length;
        System.out.println("n = " + n);
        // 计算x,y,xy,x²的总和
        double totalX = 0d;
        double totalY = 0d;
        double totalXY = 0d;
        double totalX2 = 0d;
        for (int i = 0; i < xData.length; i++) {
            totalX += xData[i];
            totalY += yData[i];
            totalXY += xData[i] * yData[i];
            totalX2 += xData[i] * xData[i];
        }
        System.out.println("totalX = " + totalX);
        System.out.println("totalY = " + totalY);
        System.out.println("totalXY = " + totalXY);
        System.out.println("totalX2 = " + totalX2);
        // 计算x均值
        double meanX = totalX / n;
        System.out.println("meanX = " + meanX);
        // 计算y均值
        double meanY = totalY / n;
        System.out.println("meanY = " + meanY);
        // 计算斜率b
        double b = (totalXY - n * meanX * meanY) / (totalX2 - n * meanX * meanX);
        System.out.println("b = " + b);
        // 计算截距a
        double a = meanY - b * meanX;
        System.out.println("a = " + a);
        // 线性方程
        System.out.printf("线性方程为: y = %.2f + %.2fxn", a, b);
        // 预测
        for (double x : preX) {
            System.out.printf("x 为%.2f时,预测y值为:%.2fn", x, a+b*x);
        }
    }

    public LeastSquaresRegression(double[] xData, double[] yData) {
        this.xData = xData;
        this.yData = yData;
    }
}
结果
n = 12
totalX = 78.0
totalY = 60200.0
totalXY = 418800.0
totalX2 = 650.0
meanX = 6.5
meanY = 5016.666666666667
b = 192.30769230769232
a = 3766.666666666667
线性方程为: y = 3766.67 + 192.31x
x 为13.00时,预测y值为:6266.67
x 为14.00时,预测y值为:6458.97
x 为15.00时,预测y值为:6651.28
x 为16.00时,预测y值为:6843.59
x 为17.00时,预测y值为:7035.90
x 为18.00时,预测y值为:7228.21
x 为19.00时,预测y值为:7420.51
x 为20.00时,预测y值为:7612.82
x 为21.00时,预测y值为:7805.13
x 为22.00时,预测y值为:7997.44
x 为23.00时,预测y值为:8189.74
x 为24.00时,预测y值为:8382.05
转载请注明:文章转载自 www.mshxw.com
本文地址:https://www.mshxw.com/it/966505.html
我们一直用心在做
关于我们 文章归档 网站地图 联系我们

版权所有 (c)2021-2022 MSHXW.COM

ICP备案号:晋ICP备2021003244-6号