python Anscombe' quartet 作业

描述https://github.com/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()

print(anascombe.groupby('dataset')['x'].agg([len,np.sum, np.mean, np.var]))
print(anascombe.groupby('dataset')['y'].agg([len,np.sum, np.mean, np.var]))
print('Icorr:' + str(anascombe['x'][0:11].corr(anascombe['y'][0:11])))
print('II:corr' + str(anascombe['x'][11:22].corr(anascombe['y'][11:22])))
print('IIIcorr:' + str(anascombe['x'][22:33].corr(anascombe['y'][22:33])))
print('IV:corr' + str(anascombe['x'][33:44].corr(anascombe['y'][33:44])))
sns.lmplot('x', 'y', anascombe[0:11])
sns.lmplot('x', 'y', anascombe[11:22])
sns.lmplot('x', 'y', anascombe[22:33])
sns.lmplot('x', 'y', anascombe[33:44])
plt.show()

结果：

          len   sum  mean   var
dataset                        
I        11.0  99.0   9.0  11.0
II       11.0  99.0   9.0  11.0
III      11.0  99.0   9.0  11.0
IV       11.0  99.0   9.0  11.0
          len    sum      mean       var
dataset                                 
I        11.0  82.51  7.500909  4.127269
II       11.0  82.51  7.500909  4.127629
III      11.0  82.50  7.500000  4.122620
IV       11.0  82.51  7.500909  4.123249
Icorr:0.81642051634484
II:corr0.8162365060002427
IIIcorr:0.8162867394895982

IV:corr0.8165214368885031