高级编程技术 jupyter课后习题

最新推荐文章于 2023-03-09 22:29:37 发布

原创最新推荐文章于 2023-03-09 22:29:37 发布 · 373 阅读

CC 4.0 BY-SA版权

本文通过Python的Pandas, Seaborn等库对Anascombe数据集进行了详细的统计分析，包括计算均值、方差、相关系数，并对每个子数据集进行线性回归分析。同时展示了数据分布的散点图。

%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")

# your code here

# 计算均值
print("\n均值:")
print(anascombe.groupby('dataset')['x'].mean())
print(anascombe.groupby('dataset')['y'].mean())

# 计算方差
print("\n方差:")
print(anascombe.groupby('dataset')['x'].var())
print(anascombe.groupby('dataset')['y'].var())

# 计算相关系数
print("\n相关系数:")
print(anascombe.x.corr(anascombe.y))

# 计算线性回归
dataset = anascombe[(anascombe['dataset']=='I')].reset_index(drop=True)  
result = smf.ols('y ~ x', dataset).fit()  
print(result.summary())
dataset = anascombe[(anascombe['dataset']=='II')].reset_index(drop=True)  
result = smf.ols('y ~ x', dataset).fit()  
print(result.summary()) 
dataset = anascombe[(anascombe['dataset']=='III')].reset_index(drop=True)  
result = smf.ols('y ~ x', dataset).fit()  
print(result.summary()) 
dataset = anascombe[(anascombe['dataset']=='IV')].reset_index(drop=True)  
result = smf.ols('y ~ x', dataset).fit()  
print(result.summary())

以数据集1为例

# your code here
sns.set(style='whitegrid')      
sns.FacetGrid(anascombe, col="dataset", hue="dataset", size=3)  .map(plt.scatter, 'x', 'y')