1. sciruby-notebooks
  2. Data Analysis
  3. Categorical Data
  4. examples
In [1]:
require 'daru'
Out[1]:
true
In [2]:
require 'open-uri'
content = open('https://d37djvu3ytnwxt.cloudfront.net/asset-v1:MITx+15.071x_3+1T2016+type@asset+block/WHO.csv')
df = Daru::DataFrame.from_csv content
df = df.at 0..6
df.first
Out[2]:
Daru::DataFrame(10x7)
Country Region Population Under15 Over60 FertilityRate LifeExpectancy
0 Afghanistan Eastern Mediterranean 29825 47.42 3.82 5.4 60
1 Albania Europe 3162 21.33 14.93 1.75 74
2 Algeria Africa 38482 27.42 7.17 2.83 73
3 Andorra Europe 78 15.2 22.86 82
4 Angola Africa 20821 47.58 3.84 6.1 51
5 Antigua and Barbuda Americas 89 25.96 12.35 2.12 75
6 Argentina Americas 41087 24.42 14.97 2.2 76
7 Armenia Europe 2969 20.34 14.06 1.74 71
8 Australia Western Pacific 23050 18.95 19.46 1.89 82
9 Austria Europe 8464 14.51 23.52 1.44 81
In [3]:
df.index = Daru::CategoricalIndex.new df['Region'].to_a
Out[3]:
#<Daru::CategoricalIndex(194): {Eastern Mediterranean, Europe, Africa, Europe, Africa, Americas, Americas, Europe, Western Pacific, Europe, Europe, Americas, Eastern Mediterranean, South-East Asia, Americas, Europe, Europe, Americas, Africa, South-East Asia ... Africa}>
In [4]:
df.first 5
Out[4]:
Daru::DataFrame(5x7)
Country Region Population Under15 Over60 FertilityRate LifeExpectancy
Eastern Mediterranean Afghanistan Eastern Mediterranean 29825 47.42 3.82 5.4 60
Europe Albania Europe 3162 21.33 14.93 1.75 74
Africa Algeria Africa 38482 27.42 7.17 2.83 73
Europe Andorra Europe 78 15.2 22.86 82
Africa Angola Africa 20821 47.58 3.84 6.1 51

Say we want to know about regions as a whole. So let's index our dataset by 'Region' vector.

List all regions

In [5]:
df.index.categories
Out[5]:
["Eastern Mediterranean", "Europe", "Africa", "Americas", "Western Pacific", "South-East Asia"]

Let's find out how many countries lie in Africa region.

In [6]:
df.row['Africa'].size
Out[6]:
46

Finding out the mean life expectancy of europe is as easy as-

In [7]:
df.row['Europe']['LifeExpectancy'].mean
Out[7]:
76.73584905660377

Let's see the maximum life expectancy of South-East Asia

In [8]:
df.row['South-East Asia']['LifeExpectancy'].max
Out[8]:
77

Set see the countries in Europe that top the list of LIfeExpectancy

In [9]:
df.row['Europe'].sort(['LifeExpectancy'], ascending: false).first 5
Out[9]:
Daru::DataFrame(5x7)
Country Region Population Under15 Over60 FertilityRate LifeExpectancy
Europe San Marino Europe 31 14.04 26.97 83
Europe Switzerland Europe 7997 14.79 23.25 1.51 83
Europe Andorra Europe 78 15.2 22.86 82
Europe France Europe 63937 18.26 23.82 1.98 82
Europe Iceland Europe 326 20.71 17.62 2.11 82

Lets see countries in South-East Asia that have high FertilityRate

In [10]:
df.row['South-East Asia'].sort(['FertilityRate']).row.at -10..-1
Out[10]:
Daru::DataFrame(10x7)
Country Region Population Under15 Over60 FertilityRate LifeExpectancy
South-East Asia Myanmar South-East Asia 52797 25.28 8.15 1.98 65
South-East Asia Democratic People's Republic of Korea South-East Asia 24763 21.98 12.74 2 69
South-East Asia Bangladesh South-East Asia 155000 30.57 6.89 2.24 70
South-East Asia Maldives South-East Asia 338 29.03 6.65 2.31 77
South-East Asia Bhutan South-East Asia 742 28.53 6.9 2.32 67
South-East Asia Sri Lanka South-East Asia 21098 25.15 12.4 2.35 75
South-East Asia Indonesia South-East Asia 247000 29.27 7.86 2.4 69
South-East Asia Nepal South-East Asia 27474 35.58 7.65 2.5 68
South-East Asia India South-East Asia 1240000 29.43 8.1 2.53 65
South-East Asia Timor-Leste South-East Asia 1114 46.33 5.16 6.11 64
In [ ]: