Data
« HomeThe specific fingerprints and their subsequent comparisons in the research paper, available for download and in the interactive tools on this website, were based on the national Input-Output tables of the World Input-Output Database (WIOD) 2013 Release (Timmer, Dietzenbacher, Los, Stehrer, de Vries 2015). For more information about WIOD and to download the national Input-Output tables, click here.
The original WIOD13 dataset covers 40 countries spanning 17 years, covering 35 sectors. However, the Household sector found in WIOD13 is excluded in the fingerprinting data due to missing data. Note that the Tab-character is used throughout as a value separator.
Note that the interactive tools also provide possibilities to download the specific data that is being visualized, either as individual tables or (through an API call) as a json-encoded file.
Complete dataset (zip package)
The following zip package contains the complete dataset for this project:
fingerprinting_complete_dataset.zipSee the _FILES_DESCR.txt in this zip file for a description of all files included in this package. For more detail about the individual files (and for downloading them separately), see below.
Reference files
_countries.txt - List of countries. 1st column: country name; 2nd column: ISO3 abbreviaton
_sectors.txt - List of economic sectors. 1st column: index number; 2nd column: sector name (shortened). Sector indices (1-34) correspond to the sector labels found in the WIOD national accounts (c1-c34). See article appendix for more details
Individual fingerprints (by country and year)
To access individual fingerprints, you can also use the Download function at the Visualize singular fingerprint tool. The table format of those downloads are slightly easier to read than those in the below zip package.
Zip package of all 680 fingerprints:
Each file consists of 68 rows and two columns. 1st column contains the sector and direction. E.g. 'c3_in' is the upstream (input) prominence of the Food sector (c3), and 'c12_out' is the downstream (output) prominence of the Metal sector (c12). 2nd column contains the prominence for that sector (i.e. up- or downstream depending on 'in' or 'out' in the first column).
Note that 27 of the country-year fingerprints were excluded in the analyses done the paper. See section 2.3 and Appendix D.1 in the supplementary information.
Eigenvalue diagnostics data
Zip package of individual eigenvalue diagnostics for each country-year (json format):
Raw Z and M matrices for each country and year
Zip package containing all Z and M matrices for each country and year, extracted from the National IO tables Excel sheet available for download from Groningen Growth and Development Centre (GGDC). Provided here under permission from GGDC (see attached README_LICENSE.txt file), this zip package also contains a simple R script to demonstrate how individual country-year fingerprints are created from these raw Z and M matrices:
Distance/dissimilarity matrix (from article)
eucl_all680.txt - File containing dissimilarity measures (Euclidean distances) for all pairs of the 680 fingerprints. Symmetric matrix with rows- and column labels.
eucl_all653.txt - File containing dissimilarity measures (Euclidean distances) for all pairs of the 653 fingerprints that were used in the first case study. Symmetric matrix with rows- and column labels.
eucl_all680_exports.txt - File containing dissimilarity measures (Euclidean distances) for all pairs of the 680 gross sectorial export vectors (E). Symmetric matrix with rows- and column labels.
Subsets (from article)
Zip package for the clusters presented in the first case study in the article:
This zip contains files with the average fingerprint statistics and country-year memberships for all partitions mentioned in the research paper. See _FILES_DESCR.txt for more details.
National longitudinal transformation data
Zip package with longitudinal transformation data for each country:
Years in first column and three columns of data. 2nd column (cpStart): structural dissimilarity between current year and 1995. 3rd column (annualChange): structural dissimilarity between current year and previous year. 4th column (cumChange): cumulative annula change, i.e. sum of all annual changes from 1995 to the current year.