January Updates
January 31, 2026
In this post I will provide an update to WiNDC activities in January 2026. I am going to work to provide, at least, monthly updates on WiNDC development and activities.
Data Release
This month saw the release of WiNDC data version 4.2.0. This update sees the inclusion of 2024 data into the build stream.
This will be the last data release using the old data and build stream. Moving forward, I will be assembling the data in Julia. However, I don’t want to move away from GAMS entirely. I am currently building a new GAMS build stream that will use data assembled in Julia.
My goal for the new GAMS build stream is make WiNDC data more reproducible and easier to use. I will be building in functionality to download specific versions of datasets used in the build stream. I will also be providing simplified files to load the WiNDC data in GAMS. This will make it much easier for users to incorporate WiNDC data into their own models.
I expect to have this available in March 2026. If there is demand from the community, I can accelerate this timeline. Further, if you are interested in testing an early version of the new build stream, please reach out. It would be great to have some early testers to provide feedback on the new process.
WiNDC JuMP Example
I’ve written a small example to introduce Julia/JuMP. This week a community member reached out to me asking for some resources to move their team from GAMS to Julia/JuMP. I provided advice and this example was a part of that advice.
The example introduces the basics of JuMP syntax and does a mild comparison to GAMS syntax. I then detail the conversion of the GAMS transport model into Julia.
This example is intended for new JuMP users. If you are interested in moving your team towards open-source tools, please contact me. I would be happy to help you get started. I have 15 years of teaching experience and can your team working quickly.
WiNDCHousehold Updates
I have been working to simplify disaggregation process in the WiNDC Household repository. I would like to make this be a template for others to use when disaggregating data. However, it is not yet at that point.
The Household model documentation is (mostly) complete. I detail the sources of each data component and each composite parameter contains the base data sources and formula used to create them. For example, see the labor supply function.