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

In [1]:

    

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


    

    




/Users/Monica/.virtualenvs/dataanalysis/lib/python3.5/site-packages/matplotlib/__init__.py:1035: UserWarning: Duplicate key in file "/Users/Monica/.matplotlib/matplotlibrc", line #2
  (fname, cnt))

In [4]:

    

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


    

In [6]:

    

df.corr()


    

    
Out[6]:






  

    

      

      
Exposure
      
Mortality
    
  
  

    

      
Exposure
      
1.000000
      
0.926345
    
    

      
Mortality
      
0.926345
      
1.000000
    
  

In [13]:

    

df.describe()


    

    
Out[13]:






  

    

      

      
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 [9]:

    

Q1 = df['Exposure'].quantile(q=0.25)
Q1


    

    
Out[9]:





2.4900000000000002

In [10]:

    

Q2 = df['Exposure'].quantile(q=0.5)
Q2


    

    
Out[10]:





3.4100000000000001

In [11]:

    

Q3 = df['Exposure'].quantile(q=0.75)
Q3


    

    
Out[11]:





6.4100000000000001

In [14]:

    

df['Mortality'].hist(bins=5)


    

    
Out[14]:





<matplotlib.axes._subplots.AxesSubplot at 0x10ba52eb8>






    

In [15]:

    

df['Mortality'].mean()


    

    
Out[15]:





157.34444444444446

In [17]:

    

df['Mort_high'] = df['Mortality'].apply(lambda x:1 if x>=157.1 else 0)
df['Expo_high'] = df['Exposure'].apply(lambda x:1 if x>=3.41 else 0)


    

In [18]:

    

df # logistic regression , high and low mortality are classified.


    

    
Out[18]:






  

    

      

      
County
      
Exposure
      
Mortality
      
Mort_high
      
Expo_high
    
  
  

    

      
0
      
Umatilla
      
2.49
      
147.1
      
0
      
0
    
    

      
1
      
Morrow
      
2.57
      
130.1
      
0
      
0
    
    

      
2
      
Gilliam
      
3.41
      
129.9
      
0
      
1
    
    

      
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 [19]:

    

lm = linear


    

    




---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
<ipython-input-19-0fc46e0387ae> in <module>()
----> 1 lm = lm.fit(x,y)

NameError: name 'lm' is not defined

In [ ]:

    




    

In [ ]:

    

lm = lm.fit(x,y)


    

In [ ]:

    

lm.predict([50])


    

In [ ]: