In [1]:

    

from statsmodels.stats.power import tt_ind_solve_power


    

In [5]:

    

normmean = 40
normstd = 17
admean = 55
adstd = 20

ef = abs(normmean-admean)/(adstd+normstd)
ratio = 1.0
alpha = 0.05
power = 0.9

n0 = tt_ind_solve_power(effect_size=ef, alpha = alpha, power = power, ratio = ratio)
print n0


    

    




[ 10.]

In [6]:

    

import pandas as pd


    

In [21]:

    

data = pd.read_excel('/home/will/ClaudioStuff/TableData.xlsx', 'Sheet1')


    

In [26]:

    

data['HIV'] = data['HIVInfected'] == 'pos'
data['Aged'] = data['Age']>50


    

In [28]:

    

tab = pd.pivot_table(data,
                     cols = ['HIV', 'Impaired'],
                     rows = 'Aged',
                     values = 'ID',
                     aggfunc = 'count').fillna(0)


    

In [29]:

    

tab.to_excel('/home/will/ClaudioStuff/count_table.xlsx')


    

In [30]:

    

tab


    

    
Out[30]:






  

    

      
HIV
      
False
      
True 
    
    

      
Impaired
      
No
      
Yes
      
No
      
Yes
    
    

      
Aged
      

      

      

      

    
  
  

    

      
False
      
 2
      
 1
      
 2
      
 23
    
    

      
True 
      
 8
      
 4
      
 0
      
  2
    
  

In [ ]: