Skewness and kurtosis

Skewness and kurtosis

skewness

Measure symmetry

skewness >0, Explain random variables x The right side deviates more from the mean , Such data distribution is called right deviation or positive deviation .

skewness >0, Explain random variables x The left side deviates more from the mean , Such a data distribution is called left bias or negative bias .

kurtosis

Kurtosis measures skewness

The kurtosis is relatively large , Explain random variables x There are many extreme values that deviate from the mean

The kurtosis is relatively small , Explain random variables x There are few extreme values that deviate from the mean

example

numpy and pandas Be responsible for preparing the data 
matplotlib Be responsible for mapping the data 

Random data

Data conforming to normal distribution

import numpy as np
import pandas
import seaborn as sns
from scipy.stats import norm
from matplotlib import pyplot as plt

data = list(np.random.randn(10000))  #  Satisfying Gaussian distribution 10000 Number 

print(' Maximum ：', max(data))
print(' minimum value ：', min(data))
print(" mean value :", np.mean(data))
print(' Median ：', np.median(data))
print(' Four percentile ：', np.percentile(data, [25, 50, 75]))
print('（ The overall ） variance ：', np.var(data))
print(' Standard deviation ：', np.std(data))

#  Skewness and kurtosis 
s = pandas.Series(data)
bias_value = s.skew()
peak_value = s.kurt()
print(" skewness ：", bias_value)
print(" kurtosis ：", peak_value)

#  mapping 
sns.distplot(data, fit=norm)
(mu, sigma) = norm.fit(data)
plt.legend(['Normal dist. ($\mu=$ {:.2f} and $\sigma=$ {:.2f} )'.format(mu, sigma)], loc='best')
plt.ylabel('Frequency')
plt.title('Distribution')
plt.show()