（十八 - 爱码网

题目：利用安斯库姆四重奏实验中的数据进行分析，计算每个表x和y的平均值，方差，相关系数和回归直线拟合，并做出图像。

题目参考网址：

https://nbviewer.jupyter.org/github/schmit/cme193-ipython-notebooks-lecture/blob/master/Exercises.ipynb

代码：

import random
import numpy as np
import scipy as sp
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

import statsmodels.api as sm
import statsmodels.formula.api as smf

sns.set_context("talk")
anascombe=pd.read_csv('data/anscombe.csv')
anascombe.head()

def Data(data):
    return pd.Series([data['x'].mean(),data['x'].var(),data['y'].mean(),data['y'].var()],index=['x mean','x var','y mean','y var'])

datasetName=anascombe.dataset.unique()
group=anascombe.groupby(by=list(["dataset"]))
for temp in datasetName:
    print()
    data=group.get_group(temp)
    print(temp)
    print(pd.DataFrame(Data(data)))
    print("相关系数")
    print(data.corr())
    x=data['x']
    x_plus=sm.add_constant(data['x'])
    y=data['y']
    model=sm.OLS(y,x_plus)
    image=model.fit()
    print(image.params)
    y_plus=image.fittedvalues
    fig,ax=plt.subplots()
    ax.plot(x,y,'o',label='data')
    ax.plot(x,y_plus,'r-',label='OLS')
    ax.legend(loc='best')
    plt.show()

结果：

I
0
x mean 9.000000
x var 11.000000
y mean 7.500909
y var 4.127269
相关系数
x y
x 1.000000 0.816421
y 0.816421 1.000000
const 3.000091
x 0.500091
dtype: float64

（十八
II
0
x mean 9.000000
x var 11.000000
y mean 7.500909
y var 4.127629
相关系数
x y
x 1.000000 0.816237
y 0.816237 1.000000
const 3.000909
x 0.500000
dtype: float64

III
0
x mean 9.00000
x var 11.00000
y mean 7.50000
y var 4.12262
相关系数
x y
x 1.000000 0.816287
y 0.816287 1.000000
const 3.002455
x 0.499727
dtype: float64
（十八

IV
0
x mean 9.000000
x var 11.000000
y mean 7.500909
y var 4.123249
相关系数
x y
x 1.000000 0.816521
y 0.816521 1.000000
const 3.001727
x 0.499909

dtype: float64

（十八