高级编程作业 jupyter:Anscombe's quartet

作业题目如下：

第一题：
代码如下：

import numpy as np
import scipy.stats
import pandas
import matplotlib.pyplot as plt
import seaborn as sns
import statsmodels.api as sm

def get_var(X):
    length = len(X)
    average = 0
    for x in X:
        average = average + x
    average = average / length
    variance = 0
    for x in X:
        variance = variance + (x - average)**2
    return variance/(length - 1)   # 这里分母为n-1


#问题一

ans = sns.load_dataset("anscombe") #题目要求的数据集

Xarray = []
Yarray = []
i = 0
while i < 4:
    sub_x = ans.x.values[i*11:i*11 + 10] #数据6个一组
    Xarray.append(sub_x)
    i = i + 1
i = 0
while i < 4:
    sub_y = ans.y.values[i*11:i*11 + 10] #数据6个一组
    Yarray.append(sub_y)
    i = i + 1

i = 0
while i < 4:
    mean = np.mean(Xarray[i])
    print("the mean of the " + str(i) + " x is "+str(mean))
    variance = get_var(Xarray[i])
    print("the variance of the " + str(i) + " x is " + str(variance))
    i = i + 1
i = 0
while i < 4:
    mean = np.mean(Yarray[i])
    print("the mean of the " + str(i) + " y is "+str(mean))
    variance = get_var(Yarray[i])
    print("the variance of the " + str(i) + " y is " + str(variance))
    i = i + 1

for i in range(0, 4):
    coefficient = scipy.stats.pearsonr(Xarray[i], Yarray[i])[0]
    print("the coefficient of "+str(i) + " is "+ str(coefficient))

for i in range(0, 4):
    model = sm.OLS(Yarray[i], sm.add_constant(Xarray[i]))
    coefficient2 = model.fit().params
    print("the fuction of "+str(i)+" is : y = " + str (coefficient2[1]) + "x" + "+" + str(coefficient2[0]))

运行结果如下：

第二题代码如下：

pic = sns.FacetGrid(ans, col = "dataset" )
pic.map(plt.scatter, 'x', 'y')
plt.show()

运行结果如图：