In [3]:

    

from __future__ import print_function
import os
import pandas as pd
from statsmodels.tsa import stattools
%matplotlib inline
from matplotlib import pyplot as plt


    

In [5]:

    

#read the data from into a pandas.DataFrame
air_miles = pd.read_csv('datasets/us-airlines-monthly-aircraft-miles-flown.csv')
air_miles.index = air_miles.Month


    

In [7]:

    

#Let's see first 10 rows of it
air_miles.head(10)


    

    
Out[7]:







  

    

      

      
Month
      
U.S. airlines: monthly aircraft miles flown (Millions) 1963 -1970
    
    

      
Month
      

      

    
  
  

    

      
1963-01
      
1963-01
      
6827.0
    
    

      
1963-02
      
1963-02
      
6178.0
    
    

      
1963-03
      
1963-03
      
7084.0
    
    

      
1963-04
      
1963-04
      
8162.0
    
    

      
1963-05
      
1963-05
      
8462.0
    
    

      
1963-06
      
1963-06
      
9644.0
    
    

      
1963-07
      
1963-07
      
10466.0
    
    

      
1963-08
      
1963-08
      
10748.0
    
    

      
1963-09
      
1963-09
      
9963.0
    
    

      
1963-10
      
1963-10
      
8194.0
    
  

In [8]:

    

#Let's rename the 2nd column
air_miles.rename(columns={'U.S. airlines: monthly aircraft miles flown (Millions) 1963 -1970':\
                          'Air miles flown'
                         },
                inplace=True
                )


    

In [10]:

    

#Plot the time series of air miles flown
fig = plt.figure(figsize=(5.5, 5.5))
ax = fig.add_subplot(1,1,1)
air_miles['Air miles flown'].plot(ax=ax)
ax.set_title('Monthly air miles flown during 1963 - 1970')
plt.savefig('plots/ch2/B07887_02_13.png', format='png', dpi=300)


    

In [19]:

    

adf_result = stattools.adfuller(air_miles['Air miles flown'], autolag='AIC')


    

In [20]:

    

print('p-val of the ADF test in air miles flown:', adf_result[1])


    

    




p-val of the ADF test in air miles flown: 0.994502281123