In [1]:

    

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


    

In [115]:

    

Data = pd.read_csv('DatosBancoMundial5.csv')
datos = Data.values[:,4:].astype(np.float64)
for i in range(5):
    datos[i,:] = datos[i,:] - np.mean(datos[i,:])


    

In [103]:

    

Data


    

    
Out[103]:







  

    

      

      
Time
      
Time Code
      
Series Name
      
Series Code
      
Colombia [COL]
      
Latin America & Caribbean [LCN]
      
Afghanistan [AFG]
      
Albania [ALB]
      
Algeria [DZA]
      
Angola [AGO]
      
...
      
Uruguay [URY]
      
Uzbekistan [UZB]
      
Vanuatu [VUT]
      
Venezuela, RB [VEN]
      
Vietnam [VNM]
      
West Bank and Gaza [PSE]
      
World [WLD]
      
Yemen, Rep. [YEM]
      
Zambia [ZMB]
      
Zimbabwe [ZWE]
    
  
  

    

      
0
      
2016
      
YR2016
      
Total tax rate (% of commercial profits)
      
IC.TAX.TOTL.CP.ZS
      
69.800000
      
45.821212
      
48.300000
      
36.500000
      
65.600000
      
48.000000
      
...
      
41.800000
      
38.100000
      
8.500000
      
64.700000
      
39.400000
      
15.300000
      
40.551323
      
33.100000
      
18.600000
      
32.800000
    
    

      
1
      
2016
      
YR2016
      
Cost of business start-up procedures (% of GNI...
      
IC.REG.COST.PC.ZS
      
7.500000
      
30.557576
      
19.900000
      
10.100000
      
11.100000
      
27.500000
      
...
      
22.500000
      
3.200000
      
44.300000
      
136.400000
      
4.600000
      
46.900000
      
25.949474
      
82.200000
      
33.700000
      
119.200000
    
    

      
2
      
2016
      
YR2016
      
Unemployment, female (% of female labor force)...
      
SL.UEM.TOTL.FE.ZS
      
13.058000
      
9.949513
      
12.368000
      
16.136000
      
19.684000
      
7.068000
      
...
      
11.206000
      
8.745000
      
6.157000
      
7.636000
      
2.107000
      
27.934000
      
6.178867
      
28.374001
      
7.706000
      
4.668000
    
    

      
3
      
2016
      
YR2016
      
Unemployment, male (% of male labor force) (mo...
      
SL.UEM.TOTL.MA.ZS
      
7.429000
      
6.727447
      
7.722000
      
16.469000
      
9.202000
      
6.186000
      
...
      
5.704000
      
8.997000
      
4.744000
      
6.400000
      
2.244000
      
24.173000
      
5.464462
      
13.073000
      
7.375000
      
5.487000
    
    

      
4
      
2016
      
YR2016
      
Ratio of female to male labor force participat...
      
SL.TLF.CACT.FM.ZS
      
72.604847
      
67.527627
      
22.960331
      
66.248318
      
23.969858
      
77.585963
      
...
      
72.744665
      
63.395067
      
76.630413
      
65.695692
      
88.722599
      
26.067661
      
67.668014
      
35.423165
      
86.347736
      
89.137578
    
  

5 rows × 227 columns

In [116]:

    

print np.shape(datos), type(datos[0,0])


    

    




(5, 223) <type 'numpy.float64'>

In [117]:

    

Cova = np.cov(datos)


    

In [118]:

    

vals, vecs = np.linalg.eig(Cova)


    

In [119]:

    

vals


    

    
Out[119]:





array([ 1153.05465923,   305.31103811,   373.56309746,    71.04135595,
           4.27342312])

In [120]:

    

vecs


    

    
Out[120]:





array([[  2.13820341e-01,   9.66977886e-01,   1.38080487e-01,
         -9.75951262e-03,   8.55346232e-03],
       [  9.76498437e-01,  -2.14321181e-01,  -9.29434474e-03,
          1.68995803e-02,  -1.20520582e-02],
       [  1.34408573e-02,  -4.16882759e-02,   1.79028285e-01,
         -7.86664227e-01,   5.89228053e-01],
       [ -5.48162881e-04,  -6.05489076e-03,   5.13482776e-02,
         -5.90983303e-01,  -8.05024923e-01],
       [  2.34671299e-02,   1.31322603e-01,  -9.72707367e-01,
         -1.77531082e-01,   6.72812081e-02]])

In [ ]: