1.WPS中

1选中数据 顶端-插入


2右键随便一个点，选择添加趋势线

3右键趋势线，设置格式，勾选公式和平方值

2.最小二乘法

import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
import math

#读取数据
def get_date(file_name,num):
    data = pd.read_csv(file_name)
    height,width = data.shape
    X_parameter = []
    Y_parameter = []
    i=1
    # 遍历数据
    for x,y in zip(data['Height'],data['Weight']):
        X_parameter.append([float(x)])
        Y_parameter.append([float(y)])
        i=i+1
        if(i>num):
            break
    return X_parameter,Y_parameter
x,y=get_date("all2.csv",20)
x1,y1=get_date("all2.csv",200)
x2,y2=get_date("all2.csv",2000)
'''输入文件路径和读取的排数'''
plt.scatter(x,y)
plt.axis()#设定坐标轴上限下限
plt.show()#读取到的点

plt.scatter(x1,y1)
plt.axis()#设定坐标轴上限下限
plt.show()#读取到的点

plt.scatter(x2,y2)
plt.axis()#设定坐标轴上限下限
plt.show()#读取到的点

#返回R平方
def computeCorrelation(X, Y):
    xBar = np.mean(X)
    yBar = np.mean(Y)
    SSR = 0
    varX = 0
    varY = 0
    for i in range(0 , len(X)):
        diffXXBar = X[i] - xBar
        diffYYBar = Y[i] - yBar
        SSR += (diffXXBar * diffYYBar)
        varX +=  diffXXBar**2
        varY += diffYYBar**2
    
    SST = math.sqrt(varX * varY)
    return math.pow((SSR / SST),2)


#返回回归方程
def regression(x,y):
    x_mean=np.mean(x)  #x平均值
    y_mean=np.mean(y)  #y平均值
    num = 0.0    #分子∑
    d=0.0    #分母∑
    for x_i,y_i in zip(x,y):
        num += (x_i-x_mean) *(y_i-y_mean)
        d +=(x_i-x_mean)**2
    a=num/d #根据公式得到a
    b=y_mean-a*x_mean #根据公式得到b
    fangcheng="y="+str(a)+"x+"+str(b)
    return fangcheng,a,b

fangcheng,a,b=regression(x, y)
fangcheng1,a1,b1=regression(x1, y1)
fangcheng2,a2,b2=regression(x2, y2)
print ("方程式1：",fangcheng,"  R²=",computeCorrelation(x, y))
print ("方程式2：",fangcheng1,"  R²=",computeCorrelation(x1, y1))
print ("方程式3：",fangcheng2,"  R²=",computeCorrelation(x2, y2))


y_hat=a*x+b  #回归方程
plt.scatter(x,y)  #描点
plt.plot(x,y_hat,color='r')  #画线
plt.axis()   #设置坐标上下限
plt.show()  #显示

y1_hat=a1*x1+b1  #回归方程
plt.scatter(x1,y1)  #描点
plt.plot(x1,y1_hat,color='r')  #画线
plt.axis()   #设置坐标上下限
plt.show()  #显示

y2_hat=a2*x2+b2  #回归方程
plt.scatter(x2,y2)  #描点
plt.plot(x2,y2_hat,color='r')  #画线
plt.axis()   #设置坐标上下限
plt.show()  #显示

3.skleran

import pandas as pd
from sklearn.linear_model import LinearRegression
import matplotlib.pyplot as plt
import numpy as np
#读取数据
def get_date(file_name,num):
    data = pd.read_csv(file_name)
    height,width = data.shape
    X_parameter = []
    Y_parameter = []
    i=1
    # 遍历数据
    for x,y in zip(data['Height'],data['Weight']):
        X_parameter.append([float(x)])
        Y_parameter.append([float(y)])
        i=i+1
        if(i>num):
            break
    return X_parameter,Y_parameter
dataSet_x,dataSet_y=get_date("all2.csv",20)
dataSet_x1,dataSet_y1=get_date("all2.csv",200)
dataSet_x2,dataSet_y2=get_date("all2.csv",2000)

#regr为回归过程，fit(x,y)进行回归
regr = LinearRegression().fit(dataSet_x, dataSet_y)
#输出R的平方
print('y=',regr.coef_,'x+',regr.intercept_)
print(regr.score(dataSet_x2, dataSet_y2))
plt.scatter(dataSet_x, dataSet_y,  color='black')
#用predic预测，这里预测输入x对应的值，进行画线
plt.plot(dataSet_x, regr.predict(dataSet_x), color='red', linewidth=1)
plt.show()


#regr为回归过程，fit(x,y)进行回归
regr = LinearRegression().fit(dataSet_x1, dataSet_y1)
#输出R的平方
print('y=',regr.coef_,'x+',regr.intercept_)
print(regr.score(dataSet_x1, dataSet_y1))
plt.scatter(dataSet_x1, dataSet_y1,  color='black')
#用predic预测，这里预测输入x对应的值，进行画线
plt.plot(dataSet_x1, regr.predict(dataSet_x1), color='red', linewidth=1)
plt.show()



#regr为回归过程，fit(x,y)进行回归
regr = LinearRegression().fit(dataSet_x2, dataSet_y2)
#输出R的平方
print('y=',regr.coef_,'x+',regr.intercept_)
print(regr.score(dataSet_x2, dataSet_y2))
plt.scatter(dataSet_x2, dataSet_y2,  color='black')
#用predic预测，这里预测输入x对应的值，进行画线
plt.plot(dataSet_x2, regr.predict(dataSet_x2), color='red', linewidth=1)
plt.show()

4.WPS结果

5.最小二乘法结果

6.skleran线性回归结果

参考链接：https://blog.csdn.net/playgoon2/article/details/77162219
结论：三种方法结果一致

原文链接:https://blog.csdn.net/qq_34591921/article/details/104842519