Weak14 Jupyter homework

%matplotlib inline

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")

Anscombe’s quartet
Anscombe’s quartet comprises of four datasets, and is rather famous. Why? You’ll find out in this exercise.

anascombe = pd.read_csv('data/anscombe.csv')
anascombe.head()

Part 1
For each of the four datasets:
Compute the mean and variance of both x and y

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

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

Compute the correlation coefficient between x and y

for i in range(4):
    X = anascombe.x[0+11*i:11+11*i].values
    Y = anascombe.y[0+11*i:11+11*i].values
    #形成的是2*2的相关系数矩阵
    print("the correlation of dataset%d is %f"%(i+1,np.corrcoef(X,Y)[0][1]))

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

for i in range(4):
    X = anascombe.x[0+11*i:11+11*i].values#11个样本点
    Y = anascombe.y[0+11*i:11+11*i].values
    train = sm.add_constant(X)#样本集左侧加上一列1，构成12维
    model = sm.OLS(Y,train).fit()
    print("The linear regression line of dataset%d is y = %f + %fx"%(i+1,model.params[0],model.params[1]))

Part 2
Using Seaborn, visualize all four datasets.
hint: use sns.FacetGrid combined with plt.scatter

m = sns.FacetGrid(anascombe, col="dataset")    
m.map(plt.scatter, "x","y")