In [1]:

    

import pandas as pd
%matplotlib inline
import matplotlib.pyplot as plt # package for doing plotting (necessary for adding the line)
import statsmodels.formula.api as smf # package we'll be using for linear regression
import matplotlib.pyplot as plt
import matplotlib
import pandas as pd
import numpy as np
plt.style.use('ggplot')
import dateutil.parser
import math
import random
import matplotlib.ticker as plticker


matplotlib.rcParams['ps.fonttype'] = 42


    

In [2]:

    

df = pd.read_csv("data/hanford.csv")
df.head()


    

    
Out[2]:






  

    

      

      
County
      
Exposure
      
Mortality
    
  
  

    

      
0
      
Umatilla
      
2.49
      
147.1
    
    

      
1
      
Morrow
      
2.57
      
130.1
    
    

      
2
      
Gilliam
      
3.41
      
129.9
    
    

      
3
      
Sherman
      
1.25
      
113.5
    
    

      
4
      
Wasco
      
1.62
      
137.5
    
  

In [16]:

    

df.describe()


    

    
Out[16]:






  

    

      

      
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 [17]:

    

df.median()


    

    
Out[17]:





Exposure       3.41
Mortality    147.10
dtype: float64

In [19]:

    

rang= df['Mortality'].max() - df['Mortality'].min()
rang


    

    
Out[19]:





96.800000000000011

In [20]:

    

iqr_m = df['Mortality'].quantile(q=0.75)- df['Mortality'].quantile(q=0.25)
iqr_m


    

    
Out[20]:





47.800000000000011

In [27]:

    

iqr_e = df['Exposure'].quantile(q=0.75)- df['Exposure'].quantile(q=0.25)
iqr_e


    

    
Out[27]:





3.9199999999999999

In [22]:

    

UAL_m= (iqr_m*1.5) + df['Mortality'].quantile(q=0.75)
UAL_m


    

    
Out[22]:





249.60000000000002

In [28]:

    

UAL_e= (iqr_m*1.5) + df['Exposure'].quantile(q=0.75)
UAL_e


    

    
Out[28]:





78.110000000000014

In [29]:

    

LAL_m= df['Mortality'].quantile(q=0.25) - (iqr_e*1.5)  
LAL_m


    

    
Out[29]:





124.22

In [30]:

    

LAL_e= df['Exposure'].quantile(q=0.25) - (iqr_e*1.5)  
LAL_e


    

    
Out[30]:





-3.3899999999999997

In [25]:

    

len(df[df['Mortality']> UAL_m])


    

    
Out[25]:





0

In [31]:

    

len(df[df['Exposure']> UAL_e])


    

    
Out[31]:





0

In [26]:

    

len(df[df['Mortality']< LAL_m])


    

    
Out[26]:





0

In [32]:

    

len(df[df['Mortality'] > UAL_m])


    

    
Out[32]:





0

In [8]:

    

df.corr()


    

    
Out[8]:






  

    

      

      
Exposure
      
Mortality
    
  
  

    

      
Exposure
      
1.000000
      
0.926345
    
    

      
Mortality
      
0.926345
      
1.000000
    
  

In [10]:

    

lm = smf.ols(formula="Mortality~Exposure",data=df).fit()
lm.params


    

    
Out[10]:





Intercept    114.715631
Exposure       9.231456
dtype: float64

In [11]:

    

def exposure_predict(exposure):
    df = pd.read_csv("data/hanford.csv")
    lm = smf.ols(formula="Mortality~Exposure",data=df).fit()
    mortality = exposure * lm.params.Exposure + lm.params.Intercept
    return mortality


    

In [13]:

    

intercept, slope = lm.params


    

In [14]:

    

df.plot(kind="scatter",x="Exposure",y="Mortality")
plt.plot(df["Exposure"],slope*df["Exposure"]+intercept,"-",color="red")


    

    
Out[14]:





[<matplotlib.lines.Line2D at 0x8076690>]






    

In [15]:

    

exposure_predict(10)


    

    
Out[15]:





207.03019352841986

In [ ]: