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

In [21]:

    

import pandas as pd
%matplotlib inline
import numpy as np
from sklearn.linear_model import LogisticRegression
import statsmodels.formula.api as smf


    

In [9]:

    

df = pd.read_csv("hanford.csv")


    

In [43]:

    

df.describe()
df.corr()


    

    
Out[43]:






  

    

      

      
Exposure
      
Mortality
    
  
  

    

      
Exposure
      
1.000000
      
0.926345
    
    

      
Mortality
      
0.926345
      
1.000000
    
  

In [32]:

    

# I could define "high exposure" as 1.5 x IQR, which would be: Q3-Q1, or 6.41-2.49
high_exposure = 4.08*1.5


    

In [33]:

    

df['Exposure'].describe()


    

    
Out[33]:





count     9.000000
mean      4.617778
std       3.491192
min       1.250000
25%       2.490000
50%       3.410000
75%       6.410000
max      11.640000
Name: Exposure, dtype: float64

In [ ]:

    




    

In [ ]:

    




    

In [40]:

    

lm = smf.ols(formula="Mortality~Exposure",data=df).fit() #notice the formula regresses Y on X (Y~X)


    

In [41]:

    

intercept, slope = lm.params


    

In [42]:

    

lm.params


    

    
Out[42]:





Intercept    114.715631
Exposure       9.231456
dtype: float64

In [ ]:

    

#y=mx+b


    

In [ ]: