GitHub

Imports

JCGE provides two import paths: data ingestion and MPSGE model conversion. The goal is to make external data and legacy model structures reusable within the JCGE ecosystem.

Data import (IO/SAM)

JCGEImportData defines a canonical CSV schema and helpers to build a SAM from IO tables. It is designed for adapters that extract data from external sources like Eurostat or GTAP.

Key outputs:

  • sam.csv
  • sets.csv
  • optional labels, subsets, mappings, and parameters
  1. Extract raw IO/SAM data from a source dataset.
  2. Map accounts into the canonical schema.
  3. Generate sets.csv and sam.csv.
  4. Validate that labels and mappings align.

Use the Calibration guide once the canonical files are in place.

MPSGE import

JCGEImportMPSGE converts an MPSGE.jl model object to a JCGE RunSpec. This is a converter and conformance bridge, not a competing authoring format.

Typical flow:

using MPSGE, JCGEImportMPSGE
m = MPSGEModel()
# build MPSGE model...
run_spec = import_mpsge(m)

When the source model is complementarity-based, the importer emits MCP blocks so the model can be solved with PATHSolver.

Choosing an import path

  • Use JCGEImportData when you are starting from data and building new models.
  • Use JCGEImportMPSGE when you need to translate an existing MPSGE model.

Next steps