标签：order   size   chapter   -o   frame   tput   eth   png   tps   

Chapter 5 - Basic Math and Statistics

Segment 6 - Delving into non-parametric methods using pandas and scipy

import numpy as np
import pandas as pd

import matplotlib.pyplot as plt
import seaborn as sb
from pylab import rcParams

import scipy
from scipy.stats import spearmanr

%matplotlib inline
rcParams[‘figure.figsize‘] = 14, 7
plt.style.use(‘seaborn-whitegrid‘)

The Spearman Rank Correlation

address = ‘~/Data/mtcars.csv‘

cars = pd.read_csv(address)
cars.columns = [‘car_names‘,‘mpg‘,‘cyl‘,‘disp‘, ‘hp‘, ‘drat‘, ‘wt‘, ‘qsec‘, ‘vs‘, ‘am‘, ‘gear‘, ‘carb‘]

cars.head()

sb.pairplot(cars)

X = cars[[‘cyl‘,‘vs‘,‘am‘,‘gear‘]]
sb.pairplot(X)

cyl = cars[‘cyl‘]
vs = cars[‘vs‘]
am = cars[‘am‘]
gear = cars[‘gear‘]

spearmanr_coefficient, p_value = spearmanr(cyl,vs)

print(‘Spearman Rank Correlation Coefficient %0.3f‘ % (spearmanr_coefficient))

Spearman Rank Correlation Coefficient -0.814

spearmanr_coefficient, p_value = spearmanr(cyl,am)

print(‘Spearman Rank Correlation Coefficient %0.3f‘ % (spearmanr_coefficient))

Spearman Rank Correlation Coefficient -0.522

spearmanr_coefficient, p_value = spearmanr(cyl,gear)

print(‘Spearman Rank Correlation Coefficient %0.3f‘ % (spearmanr_coefficient))

Spearman Rank Correlation Coefficient -0.564

Chi-square test for independence

table = pd.crosstab(cyl, am)

from scipy.stats import chi2_contingency
chi2, p, dof, expected = chi2_contingency(table.values)
print(‘Chi-square statistic %0.3f p_value %0.3f‘ % (chi2,p))

Chi-square statistic 8.741 p_value 0.013

table = pd.crosstab(cyl, vs)

from scipy.stats import chi2_contingency
chi2, p, dof, expected = chi2_contingency(table.values)
print(‘Chi-square statistic %0.3f p_value %0.3f‘ % (chi2,p))

Chi-square statistic 21.340 p_value 0.000

table = pd.crosstab(cyl, gear)

from scipy.stats import chi2_contingency
chi2, p, dof, expected = chi2_contingency(table.values)
print(‘Chi-square statistic %0.3f p_value %0.3f‘ % (chi2,p))

Chi-square statistic 18.036 p_value 0.001

Python for Data Science - Delving into non-parametric methods using pandas and scipy

标签：order   size   chapter   -o   frame   tput   eth   png   tps   

原文地址：https://www.cnblogs.com/keepmoving1113/p/14285316.html