Jupyter练习题

按照题目所给代码：

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()

Part 1：

• Compute the mean and variance of both x and y

代码：

print(anascombe.groupby('dataset')['x','y'].mean())  
print(anascombe.groupby('dataset')['x','y'].var())

结果：

• Compute the correlation coefficient between x and y

代码：

print(anascombe.cov()['x']['y'] / ((anascombe['x'].var() * anascombe['y'].var()) ** 0.5))

结果：

• Compute the linear regression line: y=β0+β1x+ϵy=β0+β1x+ϵ (hint: use statsmodels and look at the Statsmodels notebook)

代码:

linear_regression_line = smf.ols('y ~ x', anascombe).fit()  
print(linear_regression_line.summary())

结果：

Part 2：

代码：

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

结果：