Apply logistic regression to categorize whether a county had high mortality rate due to contamination

In [10]:

    

import pandas as pd
%matplotlib inline
import numpy as np
from sklearn.linear_model import LogisticRegression
import matplotlib.pyplot as plt


    

In [2]:

    

df = pd.read_csv('../data/hanford.csv')


    

In [3]:

    

df.describe()


    

    
Out[3]:






  

    

      

      
Exposure
      
Mortality
    
  
  

    

      
count
      
9.000000
      
9.000000
    
    

      
mean
      
4.617778
      
157.344444
    
    

      
std
      
3.491192
      
34.791346
    
    

      
min
      
1.250000
      
113.500000
    
    

      
25%
      
2.490000
      
130.100000
    
    

      
50%
      
3.410000
      
147.100000
    
    

      
75%
      
6.410000
      
177.900000
    
    

      
max
      
11.640000
      
210.300000
    
  

In [5]:

    

df.plot(kind = 'scatter', x = 'Exposure', y = 'Mortality')


    

    
Out[5]:





<matplotlib.axes._subplots.AxesSubplot at 0x10cf1b7b8>






    

In [12]:

    

df['Mortality'].mean() + 3 * df['Mortality'].std()


    

    
Out[12]:





261.71848336217533

In [13]:

    

df['Exposure'].mean() + 3 * df['Exposure'].std()


    

    
Out[13]:





15.09135285985935

In [16]:

    

high_exposure = df['Exposure'].median()


    

In [17]:

    

high_mortality = df['Mortality'].median()


    

In [18]:

    

df['Mort_high'] = df['Mortality'].apply(lambda x:1 if x >= high_mortality else 0)
df['Expo_high'] = df['Mortality'].apply(lambda x:1 if x >= high_mortality else 0)


    

In [19]:

    

df


    

    
Out[19]:






  

    

      

      
County
      
Exposure
      
Mortality
      
Mort_high
      
Expo_high
    
  
  

    

      
0
      
Umatilla
      
2.49
      
147.1
      
1
      
1
    
    

      
1
      
Morrow
      
2.57
      
130.1
      
0
      
0
    
    

      
2
      
Gilliam
      
3.41
      
129.9
      
0
      
0
    
    

      
3
      
Sherman
      
1.25
      
113.5
      
0
      
0
    
    

      
4
      
Wasco
      
1.62
      
137.5
      
0
      
0
    
    

      
5
      
HoodRiver
      
3.83
      
162.3
      
1
      
1
    
    

      
6
      
Portland
      
11.64
      
207.5
      
1
      
1
    
    

      
7
      
Columbia
      
6.41
      
177.9
      
1
      
1
    
    

      
8
      
Clatsop
      
8.34
      
210.3
      
1
      
1
    
  

In [20]:

    

lm = LogisticRegression()
data = np.asarray(df[['Mort_high', 'Expo_high']]) #data is a list of lists
x = data[:,1:] # the second colon is in case you have more than one input in your prediction model
y = data[:,0]
lm.fit(x,y)
lm.score(x,y)  #This gives you r^2, or coefficient of determination
slope = lm.coef_[0] #the coefficients are a list (here, we’re just using one) of the slopes for each x value
intercept = lm.intercept_


    

In [22]:

    

df.plot(kind = 'scatter', x = 'Exposure', y = 'Mortality')
plt.plot(df['Expo_high'], slope * df['Mort_high'] + intercept, '-')


    

    
Out[22]:





[<matplotlib.lines.Line2D at 0x10cf720b8>]






    

In [23]:

    

lm.predict(50)


    

    
Out[23]:





array([1])

In [ ]: