第十四周作业

可以看到这是四个数据集，按照要求依次计算，最后绘制散点图。

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
anascombe = pd.read_csv('anscombe.csv')
anascombe.head()
print(anascombe)

x1 = anascombe.x[:10].values    
x2 = anascombe.x[11:21].values
x3 = anascombe.x[22:32].values
x4 = anascombe.x[33:43].values

y1 = anascombe.y[:10].values  
y2 = anascombe.y[11:21].values
y3 = anascombe.y[22:32].values
y4 = anascombe.y[33:43].values
print("\n")
print("Part 1.1")
_x1 = np.mean(x1)
_y1 = np.mean(y1)
dx1 = np.var(x1)
dy1 = np.var(y1)
print("mean of x1 : ", _x1)
print("mean of y1 : ", _y1)
print("variance of x1 : ", dx1) 
print("variance of y1 : ", dy1) 

_x2 = np.mean(x2)
_y2 = np.mean(y2)
dx2 = np.var(x2)
dy2 = np.var(y2)
print("mean of x2 : ", _x2)
print("mean of y2 : ", _y2)
print("variance of x2 : ", dx2) 
print("variance of y2 : ", dy2) 

_x3 = np.mean(x3)
_y3 = np.mean(y3)
dx3 = np.var(x3)
dy3 = np.var(y3)
print("mean of x3 : ", _x3)
print("mean of y3 : ", _y3)
print("variance of x3 : ", dx3) 
print("variance of y3 : ", dy3) 

_x4 = np.mean(x4)
_y4 = np.mean(y4)
dx4 = np.var(x4)
dy4 = np.var(y4)
print("mean of x4 : ", _x4)
print("mean of y4 : ", _y4)
print("variance of x4 : ", dx4) 
print("variance of y4 : ", dy4) 
print("\n")
print("Part 1.2")
coeff1 = np.corrcoef(x1,y1)
print("correlation coefficient between x1 and y1 is ",coeff1[0][1])

coeff2 = np.corrcoef(x2,y2)
print("correlation coefficient between x2 and y2 is ",coeff2[0][1])

coeff3 = np.corrcoef(x3,y3)
print("correlation coefficient between x3 and y3 is ",coeff3[0][1])

coeff4 = np.corrcoef(x4,y4)
print("correlation coefficient between x4 and y4 is ",coeff4[0][1])

print("\n")
print("Part 1.3")
X1 = sm.add_constant(x1)
est1 = sm.OLS(y1,X1)
est1 = est1.fit()
para1 = est1.params
print("Dataset1: y1 =", para1[0], "+", para1[1], "* x1")

X2 = sm.add_constant(x2)
est2 = sm.OLS(y2,X2)
est2 = est2.fit()
para2 = est2.params
print("Dataset2: y2 =", para2[0], "+", para2[1], "* x2")

X3 = sm.add_constant(x3)
est3 = sm.OLS(y3,X3)
est3 = est3.fit()
para3 = est3.params
print("Dataset3: y3 =", para3[0], "+", para3[1], "* x3")

X4 = sm.add_constant(x4)
est4 = sm.OLS(y4,X4)
est4 = est4.fit()
para4 = est4.params
print("Dataset4: y4 =", para4[0], "+", para4[1], "* x4")

print("\n")

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