A two-layer example: clustering of countries in respect of their export and socioeconomic data



The goal is to demonstrate usefulness of a two layer clustering. In the first layer are data about export of a country as reported by UNCTAD for year 2012 while in the second layer are socieconomic World Bank data for the same year. In the first data set are 106 attributes presenting percentage of export for specific commodities (like percentage of export of chemical products) and in the second data set are 105 attributes like percentage of population younger than 14 years or money and quasi money (M2) as % of GDP. The data describe in total 155 countries. There are a lot of unknown values in both layers.
The set of files:
export data .
socieconomic data .
names of 155 countries .
list of 106 attributes in the export data set .
list of 105 attributes in the socioeconomic data set .
list of 106 descriptions of export attributes (for human interpretation only).
list of 105 descriptions of socioeconomic attributes (for human interpretation only) .

Read tutorial describng how this files are used to obtain clusters of countries.


© 2016 LIS - Rudjer Boskovic Institute
Last modified: January 09 2017 21:21:51.