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

1. Import the necessary packages to read in the data, plot, and create a logistic regression model



In [40]:

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

2. Read in the hanford.csv file in the `data/` folder



In [41]:

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

3. Calculate the basic descriptive statistics on the data



In [42]:

    
df.describe()



In [43]:

    
iqr = df.quantile(q=0.75) - df.quantile(q=0.25)
iqr









    Out[43]:





Exposure      3.92
Mortality    47.80
dtype: float64



In [44]:

    
df.boxplot(column='Exposure', return_type='axes')









    Out[44]:





<matplotlib.axes._subplots.AxesSubplot at 0x10c430780>

4. Find a reasonable threshold to say exposure is high and recode the data



In [45]:

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



In [46]:

    
df_high = df[df['Exposure'] >= threshold]
df_high.shape









    Out[46]:





(5, 3)



In [47]:

    
df_high









    Out[47]:






  
    
      
      County
      Exposure
      Mortality
    
  
  
    
      2
      Gilliam
      3.41
      129.9
    
    
      5
      HoodRiver
      3.83
      162.3
    
    
      6
      Portland
      11.64
      207.5
    
    
      7
      Columbia
      6.41
      177.9
    
    
      8
      Clatsop
      8.34
      210.3



In [ ]:

5. Create a logistic regression model



In [61]:

    
#lr = LogisticRegression(df_high)
lm = LogisticRegression()



In [72]:

    
#X = df_high['Exposure'].values
#y = df_high['Mortality'].values

data = np.asarray(df_high[['Mortality','Exposure']])
#X = data[:,1:] # ',' = all of the rows
#y = data[:,0]

x = np.asarray(df_high[['Exposure']])
y = np.asarray(df_high['Mortality'])
#lr.fit(x, y)
x, y









    Out[72]:





(array([[  3.41],
        [  3.83],
        [ 11.64],
        [  6.41],
        [  8.34]]), array([ 129.9,  162.3,  207.5,  177.9,  210.3]))



In [ ]:

6. Predict whether the mortality rate (Cancer per 100,000 man years) will be high at an exposure level of 50



In [64]:

    
#lr.predict(50)



In [ ]:

	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

	County	Exposure	Mortality
2	Gilliam	3.41	129.9
5	HoodRiver	3.83	162.3
6	Portland	11.64	207.5
7	Columbia	6.41	177.9
8	Clatsop	8.34	210.3

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

1. Import the necessary packages to read in the data, plot, and create a logistic regression model

2. Read in the hanford.csv file in the data/ folder

3. Calculate the basic descriptive statistics on the data

4. Find a reasonable threshold to say exposure is high and recode the data

5. Create a logistic regression model

6. Predict whether the mortality rate (Cancer per 100,000 man years) will be high at an exposure level of 50

2. Read in the hanford.csv file in the `data/` folder