In [12]:

    

import pandas as pd                   # dataframe
import matplotlib.pyplot as plt       # graphs & other analysis
import matplotlib as mpl              # graphics parameters
import numpy as np                    # numerical calculations


    

In [4]:

    

df = pd.read_excel("2013_NYC_CD_MedianIncome_Recycle.xlsx")


    

In [5]:

    

df.head()


    

    
Out[5]:






  

    

      

      
CD_Name
      
MdHHIncE
      
RecycleRate
    
  
  

    

      
0
      
Battery Park City, Greenwich Village & Soho
      
119596
      
0.286771
    
    

      
1
      
Battery Park City, Greenwich Village & Soho
      
119596
      
0.264074
    
    

      
2
      
Chinatown & Lower East Side
      
40919
      
0.156485
    
    

      
3
      
Chelsea, Clinton & Midtown Business Distric
      
92583
      
0.235125
    
    

      
4
      
Chelsea, Clinton & Midtown Business Distric
      
92583
      
0.246725
    
  

In [9]:

    

%matplotlib inline
df.plot(kind='scatter',y='MdHHIncE',x='RecycleRate')
df.corr()


    

    
Out[9]:






  

    

      

      
MdHHIncE
      
RecycleRate
    
  
  

    

      
MdHHIncE
      
1.000000
      
0.884783
    
    

      
RecycleRate
      
0.884783
      
1.000000
    
  








    

In [30]:

    

def plot_correlation( ds, x, y, dotcolor, linecolor, ylim=(0,140000) ):
    plt.xlim(.05,.35)
    plt.ylim(ylim[0],ylim[1])
    plt.scatter(ds[x], ds[y], alpha=0.6, s=50, c=dotcolor) 
    plt.xlabel(x)
    plt.ylabel(y)
    
    # Correlation 
    trend_variable = np.poly1d(np.polyfit(ds[x], ds[y], 1))
    trendx = np.linspace(0, 0.35, 4)
    plt.plot(trendx, trend_variable(trendx), color=linecolor) 
    r = sp.stats.pearsonr(ds[x],ds[y])
    plt.text(trendx[3], trend_variable(trendx[3]),'r={:.3f}'.format(r[0]), color = linecolor )
    plt.tight_layout()


    

In [31]:

    

import scipy as sp
from scipy import stats


    

In [32]:

    

plot_correlation(df,'RecycleRate','MdHHIncE','b','r')


    

In [ ]: