python homework——the 14th week

Part1

1.
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
import statsmodels.formula.api as sfa

anc=pd.read_csv('anscombe.csv')
print('The mean of x is\n'+str(anc.groupby('dataset')['x'].mean()))
print('The mean of y is\n'+str(anc.groupby('dataset')['y'].mean()))
print('The variance of x is\n'+str(anc.groupby('dataset')['x'].var()))
print('The variance of y is\n'+str(anc.groupby('dataset')['y'].var()))

2.

import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
import statsmodels.formula.api as sfa

anc=pd.read_csv('anscombe.csv')
print('The correlation coefficient of the first dataset is \n'+str(anc[anc['dataset']=='I'].corr()))
print('The correlation coefficient of the second dataset is \n'+str(anc[anc['dataset']=='II'].corr()))
print('The correlation coefficient of the third dataset is \n'+str(anc[anc['dataset']=='III'].corr()))
print('The correlation coefficient of the fourth dataset is \n'+str(anc[anc['dataset']=='IV'].corr()))

3.

import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
import statsmodels.formula.api as sfa

anc=pd.read_csv('anscombe.csv')

md=[0,0,0,0]
md[0] = sfa.ols('y ~ x', anc[anc['dataset'] == 'I']).fit()
md[1] = sfa.ols('y ~ x', anc[anc['dataset'] == 'II']).fit()
md[2] = sfa.ols('y ~ x', anc[anc['dataset'] == 'III']).fit()
md[3] = sfa.ols('y ~ x', anc[anc['dataset'] == 'IV']).fit()
for i in range(4):
     print('\n')
     print(md[i].summary())

Part2

import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
import statsmodels.formula.api as sfa

anc=pd.read_csv('anscombe.csv')

vsl=sns.FacetGrid(anc,col='dataset')
vsl.map(plt.scatter,'x','y')
plt.show()