C++-Implementation of the flexible algorithm for regional and sectoral disaggregation of multi-regional input-output tables as described in:
L. Wenz, S.N. Willner, A. Radebach, R. Bierkandt, J.C. Steckel, A. Levermann
Regional and sectoral disaggregation of multi-regional input-output tables: a flexible algorithm
Economic Systems Research 27 (2015), DOI: 10.1080/09535314.2014.987731.
It includes a library for handling heterogeneous MRIO tables with up to one level of hierarchy. If you want to use it and have trouble with it just drop me an email.
Just use cmake:
mkdir build
cd build
cmake ..
to create the mrio_disaggregate binary. Compiler has to support C++11.
To use libmrio as a static library you can include libmrio.cmake from cmake.
Libmrio depends on my [https://github.com/swillner/cpp-library](CSV parser) and [https://github.com/swillner/settingsnode](Settings wrapper). These are included as subtrees.
It also uses https://github.com/jbeder/yaml-cpp.git, which you either need to have installed or get as a submodule:
git submodule update --init --recursive
The implementation optionally uses the https://github.com/Unidata/netcdf-cxx4 (e.g. package libnetcdf-c++4-dev in Ubuntu/Debian). Its use is controlled via the cmake option LIBMRIO_WITH_NETCDF (e.g. use ccmake ..).
mrio_disaggregate expects the path of a YAML control file as parameter (or - and the file is read from stdin):
-
YAML control file see example in
examples/simple. -
Proxy files CSV-files with proxy data. Column numbers depend on proxy level (as documented in the paper). First column: Year; Then columns of either region/sector name or column pairs of region/sector name and index (starting with 0) of subregion/subsector; Then value; Concluding with an optional column given the sum (only applies for GDP and population levels).
-
Formats Supported input and output formats are
csv,mrioandnetcdf.