In [11]:

    

%matplotlib inline
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

"""
Type 1 - Count & Probability for Discrete variable
Type 2 - Equal Bin Size, Count & Probability for Continuous variable
Type 3 - Normalized Count for Discrete variable
Type 4 - Equal Bin Size, Normalized Count for Continuous variable
"""


    

In [2]:

    

preg_raw = pd.read_csv("2002FemPregOut.csv")
reps_raw = pd.read_csv("2002FemRespOut.csv")


    

In [4]:

    

preg_raw = preg_raw.dropna()
preg_raw.head()


    

    
Out[4]:







  

    

      

      
caseid
      
nbrnaliv
      
babysex
      
birthwgt_lb
      
birthwgt_oz
      
prglngth
      
outcome
      
birthord
      
agepreg
      
finalwgt
    
  
  

    

      
0
      
1
      
1.0
      
1.0
      
8.0
      
13.0
      
39
      
1
      
1.0
      
3316.0
      
6448.271112
    
    

      
1
      
1
      
1.0
      
2.0
      
7.0
      
14.0
      
39
      
1
      
2.0
      
3925.0
      
6448.271112
    
    

      
2
      
2
      
3.0
      
1.0
      
9.0
      
2.0
      
39
      
1
      
1.0
      
1433.0
      
12999.542264
    
    

      
3
      
2
      
1.0
      
2.0
      
7.0
      
0.0
      
39
      
1
      
2.0
      
1783.0
      
12999.542264
    
    

      
4
      
2
      
1.0
      
2.0
      
6.0
      
3.0
      
39
      
1
      
3.0
      
1833.0
      
12999.542264
    
  

In [8]:

    

birthorder_ct = preg_raw[['birthord', 'caseid']]\
                      .groupby(['birthord'])['caseid']\
                      .agg(['count'])\
                      .sort_values(['count'], ascending=False)
print birthorder_ct
print birthorder_ct.sum()


    

    




          count
birthord       
1.0        4381
2.0        2861
3.0        1225
4.0         417
5.0         125
6.0          49
7.0          19
8.0           7
9.0           2
10.0          1
count    9087
dtype: int64

In [10]:

    

# Type 1 - Count & Probability for Discrete variable
birthorder_ct['prob'] = birthorder_ct/birthorder_ct.sum()
birthorder_ct


    

    
Out[10]:







  

    

      

      
count
      
prob
    
    

      
birthord
      

      

    
  
  

    

      
1.0
      
4381
      
0.482117
    
    

      
2.0
      
2861
      
0.314845
    
    

      
3.0
      
1225
      
0.134808
    
    

      
4.0
      
417
      
0.045890
    
    

      
5.0
      
125
      
0.013756
    
    

      
6.0
      
49
      
0.005392
    
    

      
7.0
      
19
      
0.002091
    
    

      
8.0
      
7
      
0.000770
    
    

      
9.0
      
2
      
0.000220
    
    

      
10.0
      
1
      
0.000110
    
  

In [22]:

    

birthorder_ct.plot.bar(x=birthorder_ct.index, y='prob')


    

    
Out[22]:





<matplotlib.axes._subplots.AxesSubplot at 0x10ee94890>






    

In [26]:

    

# Type 3 - Normalized Count for Discrete variable
## normalize to [0,1] range
birthorder_ct['normalized_ct'] = (birthorder_ct['count']-birthorder_ct['count'].min())/(birthorder_ct['count'].max()-birthorder_ct['count'].min())
birthorder_ct


    

    
Out[26]:







  

    

      

      
count
      
prob
      
normalized_ct
    
    

      
birthord
      

      

      

    
  
  

    

      
1.0
      
4381
      
0.482117
      
1.000000
    
    

      
2.0
      
2861
      
0.314845
      
0.652968
    
    

      
3.0
      
1225
      
0.134808
      
0.279452
    
    

      
4.0
      
417
      
0.045890
      
0.094977
    
    

      
5.0
      
125
      
0.013756
      
0.028311
    
    

      
6.0
      
49
      
0.005392
      
0.010959
    
    

      
7.0
      
19
      
0.002091
      
0.004110
    
    

      
8.0
      
7
      
0.000770
      
0.001370
    
    

      
9.0
      
2
      
0.000220
      
0.000228
    
    

      
10.0
      
1
      
0.000110
      
0.000000
    
  

In [27]:

    

birthorder_ct.plot.bar(x=birthorder_ct.index, y='normalized_ct')


    

    
Out[27]:





<matplotlib.axes._subplots.AxesSubplot at 0x10f182c50>






    

In [30]:

    

# Type 2 - Equal Bin Size, Count & Probability for Continuous variable
## by default, 10 bins
hist_vals, bin_edges = np.histogram(preg_raw['prglngth'])
print hist_vals
print bin_edges


    

    




[   2    1    0    4   24   92  387 6336 2220   21]
[  0.   5.  10.  15.  20.  25.  30.  35.  40.  45.  50.]

In [32]:

    

plt.bar(bin_edges[:-1], hist_vals, width = 1)
plt.xlim(min(bin_edges), max(bin_edges))
plt.show()


    

In [40]:

    

# Type 4 - Equal Bin Size, Normalized Count for Continuous variable
min_hist_val = min(hist_vals)
max_hist_val = max(hist_vals)
normalized_hist_vals = [(elem-min_hist_val)*1.0/(max_hist_val-min_hist_val) for elem in hist_vals]
normalized_hist_vals


    

    
Out[40]:





[0.00031565656565656568,
 0.00015782828282828284,
 0.0,
 0.00063131313131313137,
 0.003787878787878788,
 0.01452020202020202,
 0.061079545454545456,
 1.0,
 0.3503787878787879,
 0.0033143939393939395]

In [41]:

    

plt.bar(bin_edges[:-1], normalized_hist_vals, width = 1)
plt.xlim(min(bin_edges), max(bin_edges))
plt.show()