In [1]:

    

%matplotlib inline

import pandas as pd
from scipy import stats
import seaborn as sns
from stemgraphic import stem_graphic


    

In [2]:

    

!wget http://download.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Medicare-Provider-Charge-Data/Downloads/Medicare_Provider_Util_Payment_PUF_CY2012_update.zip


    

    




--2016-09-13 11:44:32--  http://download.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Medicare-Provider-Charge-Data/Downloads/Medicare_Provider_Util_Payment_PUF_CY2012_update.zip
Resolving download.cms.gov... 204.156.15.130, 204.156.15.106, 2001:5014:703::3f82:4e09, ...
Connecting to download.cms.gov|204.156.15.130|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 467801269 (446M) [application/zip]
Saving to: “Medicare_Provider_Util_Payment_PUF_CY2012_update.zip”

100%[======================================>] 467,801,269 3.82M/s   in 2m 44s  

2016-09-13 11:47:16 (2.72 MB/s) - “Medicare_Provider_Util_Payment_PUF_CY2012_update.zip” saved [467801269/467801269]

In [3]:

    

!unzip -o Medicare_Provider_Util_Payment_PUF_CY2012_update.zip


    

    




Archive:  Medicare_Provider_Util_Payment_PUF_CY2012_update.zip
  inflating: CMS_AMA_CPT_license_agreement.pdf  
  inflating: Medicare-Physician-and-Other-Supplier-PUF-SAS-Infile.sas  
  inflating: Medicare_Provider_Util_Payment_PUF_CY2012.txt  

In [4]:

    

df = pd.read_table("Medicare_Provider_Util_Payment_PUF_CY2012.txt", skiprows=[1], header=0)


    

    




/usr/local/lib/python3.4/site-packages/IPython/core/interactiveshell.py:2705: DtypeWarning:

Columns (10) have mixed types. Specify dtype option on import or set low_memory=False.

In [5]:

    

df.describe()


    

    
Out[5]:






  

    

      

      
NPI
      
LINE_SRVC_CNT
      
BENE_UNIQUE_CNT
      
BENE_DAY_SRVC_CNT
      
AVERAGE_MEDICARE_ALLOWED_AMT
      
STDEV_MEDICARE_ALLOWED_AMT
      
AVERAGE_SUBMITTED_CHRG_AMT
      
STDEV_SUBMITTED_CHRG_AMT
      
AVERAGE_MEDICARE_PAYMENT_AMT
      
STDEV_MEDICARE_PAYMENT_AMT
    
  
  

    

      
count
      
9.153272e+06
      
9.153272e+06
      
9.153272e+06
      
9.153272e+06
      
9.153272e+06
      
9.153272e+06
      
9.153272e+06
      
9.153272e+06
      
9.153272e+06
      
9.153272e+06
    
    

      
mean
      
1.499546e+09
      
2.397031e+02
      
9.119919e+01
      
1.509448e+02
      
9.628829e+01
      
6.568267e+00
      
2.981586e+02
      
2.229884e+01
      
7.490089e+01
      
1.240657e+01
    
    

      
std
      
2.878767e+08
      
5.104884e+03
      
1.137321e+03
      
2.180008e+03
      
2.033851e+02
      
4.855068e+01
      
7.913027e+02
      
1.611956e+02
      
1.610363e+02
      
4.157077e+01
    
    

      
min
      
1.003000e+09
      
6.000000e+00
      
1.100000e+01
      
1.100000e+01
      
3.333330e-05
      
0.000000e+00
      
3.333330e-05
      
0.000000e+00
      
0.000000e+00
      
0.000000e+00
    
    

      
25%
      
1.245479e+09
      
2.100000e+01
      
1.700000e+01
      
2.000000e+01
      
2.179000e+01
      
0.000000e+00
      
5.000000e+01
      
0.000000e+00
      
1.747260e+01
      
4.654928e-02
    
    

      
50%
      
1.497931e+09
      
4.500000e+01
      
3.300000e+01
      
4.200000e+01
      
6.019000e+01
      
0.000000e+00
      
1.228000e+02
      
0.000000e+00
      
4.457513e+01
      
4.978978e+00
    
    

      
75%
      
1.740553e+09
      
1.240000e+02
      
7.700000e+01
      
1.110000e+02
      
1.086000e+02
      
7.577722e-01
      
2.525000e+02
      
3.492409e+00
      
8.233045e+01
      
1.581299e+01
    
    

      
max
      
1.993000e+09
      
4.578582e+06
      
6.046340e+05
      
1.227047e+06
      
4.627252e+04
      
4.480343e+04
      
9.800000e+04
      
5.319586e+04
      
3.701232e+04
      
3.583923e+04
    
  

In [6]:

    

nebulizer = df[df.HCPCS_CODE=='94640']


    

In [7]:

    

nebulizer['AVERAGE_SUBMITTED_CHRG_AMT'].hist()


    

    
Out[7]:





<matplotlib.axes._subplots.AxesSubplot at 0x7fb2c9fb9f28>






    

In [8]:

    

sns.rugplot(nebulizer['AVERAGE_SUBMITTED_CHRG_AMT'])


    

    
Out[8]:





<matplotlib.axes._subplots.AxesSubplot at 0x7fb308f40940>






    

In [9]:

    

sns.stripplot(nebulizer['AVERAGE_SUBMITTED_CHRG_AMT'], jitter=True)


    

    
Out[9]:





<matplotlib.axes._subplots.AxesSubplot at 0x7fb2fde879e8>






    

In [10]:

    

sns.distplot(nebulizer['AVERAGE_SUBMITTED_CHRG_AMT'], kde=False)


    

    
Out[10]:





<matplotlib.axes._subplots.AxesSubplot at 0x7fb2fd4f6e48>






    

In [11]:

    

sns.distplot(nebulizer['AVERAGE_SUBMITTED_CHRG_AMT'], rug=True)


    

    
Out[11]:





<matplotlib.axes._subplots.AxesSubplot at 0x7fb2fd318c88>






    

In [12]:

    

ax = sns.distplot(df[df.HCPCS_CODE=='94640']['AVERAGE_SUBMITTED_CHRG_AMT'], fit=stats.norm)
ax.set_title('kde vs normal')


    

    
Out[12]:





<matplotlib.text.Text at 0x7fb2f5d96ac8>






    

In [13]:

    

ax = sns.distplot(df[df.HCPCS_CODE=='94640']['AVERAGE_SUBMITTED_CHRG_AMT'], fit=stats.dgamma)
ax.set_title('kde vs double gamma')


    

    
Out[13]:





<matplotlib.text.Text at 0x7fb2f5c1b908>






    

In [14]:

    

g = sns.factorplot(x='AVERAGE_SUBMITTED_CHRG_AMT', y='PROVIDER_TYPE',
                   data=nebulizer[nebulizer.AVERAGE_SUBMITTED_CHRG_AMT>0],
                   hue='MEDICARE_PARTICIPATION_INDICATOR', split=True, kind='violin',
                   orient="h", size=7, aspect=2, palette="Set3", bw=.2, cut=0)


    

In [15]:

    

sub_nebulizer = nebulizer[nebulizer.PROVIDER_TYPE.isin(['Pediatric Medicine','Pulmonary Disease', 'Family Practice','Emergency Medicine'])]


    

In [16]:

    

g = sns.factorplot(x='AVERAGE_SUBMITTED_CHRG_AMT', y='PROVIDER_TYPE',
                   data=sub_nebulizer, kind='box',
                   orient="h", size=6, aspect=3, palette="Set3")


    

In [18]:

    

stem_graphic(sub_nebulizer['AVERAGE_SUBMITTED_CHRG_AMT'], unit='$')


    

    
Out[18]:





(<matplotlib.figure.Figure at 0x7fb2f5dc3f60>,
 <matplotlib.axes._axes.Axes at 0x7fb2f641c080>)






    

In [ ]: