In [1]:

    

import pandas as pd
import scipy as sp
import numpy as np

from abtools import ABtest, LognormalModel
from abtools.plotting import qqplot, ppplot

%matplotlib inline
import matplotlib.pyplot as plt
import seaborn as sns
sns.set_style('whitegrid')


    

In [2]:

    

data = pd.read_csv('data.csv').query('revenue > 0')


    

In [3]:

    

log_data = np.log(data.revenue)


    

In [14]:

    

m = LognormalModel(data['revenue'])


    

In [15]:

    

qqplot(log_data, np.log(m.sample_ppc(1000)))


    

    
Out[15]:





[<matplotlib.axes._subplots.AxesSubplot at 0x7f612057efd0>]






    

In [17]:

    

a = data.query('abgroup == "control"')['revenue']
b = data.query('abgroup == "test"')['revenue']


    

In [22]:

    

test = ABtest(LognormalModel, [a, b], 5000)

test.test_all()


    

    




ABtest for 2 groups






    
Out[22]:





<abtools.core.abtest.ABtest at 0x7f6120390710>

In [23]:

    

test.plot()


    

In [24]:

    

test.probabilities_df


    

    
Out[24]:







  

    

      

      
group1
      
group2
      
mean
    
  
  

    

      
group1
      
-
      
0.9754
      
21.8011425844
    
    

      
group2
      
0.0246
      
-
      
22.4173097151
    
    

      
mean
      
21.8011425844
      
22.4173097151
      
-
    
  

In [ ]: