Uploaded November 2026 | Updated November 2026, 3 hours ago
Cross-Country Macroeconomic Forecasting Using Physics-Informed Neural Networks and Universal Differential Equations in Julia by Vrishank Sai Anand
PreTalx: pretalx.com/juliacon-2026/talk/AHVQ3V
Macroeconomic debt cycles evolve over time through interacting growth, inflation, and fiscal pressures. This study compares India, Sri Lanka, and Argentina using Physics-Informed Neural Networks (PINNs) and Universal Differential Equations (UDEs) to model continuous economic dynamics. By embedding economic structure into neural differential systems, we evaluate their ability to reconstruct national macroeconomic trajectories under sparse annual data conditions.
This project investigates whether continuous-time scientific machine learning methods can better represent such evolving economic systems. Building on the idea of long-term debt cycles, the study moves beyond a single country setting to compare India, Sri Lanka, and Argentina. Three economies with distinct fiscal histories and crisis trajectories. By analyzing multiple countries, the research expands its horizon from local model fitting to cross-country structural comparison. The central question is not about which model predicts better, but how embedding economic structure into learning systems affects stability, interpretability, and long-term trajectory reconstruction.
The methodology is based on Physics-Informed Neural Networks (PINNs) and Universal Differential Equations (UDEs). Unlike purely data-driven neural models, these approaches integrate differential equation structures and domain constraints directly into the learning process. This allows the model to respect economic relationships such as debt-growth feedback and dynamic fiscal adjustments while still learning unknown components from data. The implementation is carried out in Julia, leveraging its high-performance scientific computing ecosystem.
In particular, the project utilizes the DifferentialEquations.jl suite for numerical ODE solving, DiffEqFlux.jl and SciMLSensitivity.jl for neural differential equation training and adjoint-based gradient computation, and NeuralPDE.jl for physics-informed learning. Optimization is handled using Optimization.jl with ADAM and second-order solvers, while data processing and evaluation rely on DataFrames.jl and Statistics. Julia’s amazing SciML framework enables tight coupling between symbolic differential equations and neural networks, allowing us to experiment with both structural constraints and hybrid modeling. Using annual macroeconomic indicators, the study evaluates how well these hybrid systems reconstruct historical trajectories under sparse sampling conditions.
By comparing performance across three countries, the research explores how structural constraints influence validity, error accumulation, and generalization. Ultimately, this project aims to bridge macroeconomic modeling and scientific machine learning, advancing a more disciplined and interpretable approach to modeling national economic dynamics.
Cross-Country Macroeconomic Forecasting Using Physics-Informed Neural Networks and Universal Differential Equations in Julia by Vrishank Sai Anand
PreTalx: pretalx.com/juliacon-2026/talk/AHVQ3V
Macroeconomic debt cycles evolve over time through interacting growth, inflation, and fiscal pressures. This study compares India, Sri Lanka, and Argentina using Physics-Informed Neural Networks (PINNs) and Universal Differential Equations (UDEs) to model continuous economic dynamics. By embedding economic structure into neural differential systems, we evaluate their ability to reconstruct national macroeconomic trajectories under sparse annual data conditions.
This project investigates whether continuous-time scientific machine learning methods can better represent such evolving economic systems. Building on the idea of long-term debt cycles, the study moves beyond a single country setting to compare India, Sri Lanka, and Argentina. Three economies with distinct fiscal histories and crisis trajectories. By analyzing multiple countries, the research expands its horizon from local model fitting to cross-country structural comparison. The central question is not about which model predicts better, but how embedding economic structure into learning systems affects stability, interpretability, and long-term trajectory reconstruction.
The methodology is based on Physics-Informed Neural Networks (PINNs) and Universal Differential Equations (UDEs). Unlike purely data-driven neural models, these approaches integrate differential equation structures and domain constraints directly into the learning process. This allows the model to respect economic relationships such as debt-growth feedback and dynamic fiscal adjustments while still learning unknown components from data. The implementation is carried out in Julia, leveraging its high-performance scientific computing ecosystem.
In particular, the project utilizes the DifferentialEquations.jl suite for numerical ODE solving, DiffEqFlux.jl and SciMLSensitivity.jl for neural differential equation training and adjoint-based gradient computation, and NeuralPDE.jl for physics-informed learning. Optimization is handled using Optimization.jl with ADAM and second-order solvers, while data processing and evaluation rely on DataFrames.jl and Statistics. Julia’s amazing SciML framework enables tight coupling between symbolic differential equations and neural networks, allowing us to experiment with both structural constraints and hybrid modeling. Using annual macroeconomic indicators, the study evaluates how well these hybrid systems reconstruct historical trajectories under sparse sampling conditions.
By comparing performance across three countries, the research explores how structural constraints influence validity, error accumulation, and generalization. Ultimately, this project aims to bridge macroeconomic modeling and scientific machine learning, advancing a more disciplined and interpretable approach to modeling national economic dynamics.
 is a fully Julia-based idealized atmospheric flow solver, primarily developed for gravity wave research. The model is highly efficient and integrates the Boussinesq, pseudo-incompressible, and compressible equations over arbitrary topography. Either high-resolution, wave-resolving simulations can be conducted, or simulations with parameterized gravity waves using the parameterization scheme MS-GWaM, where both transience and horizontal propagation of gravity waves are accounted for.
[PinCFlow.jl](https://github.com/Atmospheric-Dynamics-GUF/PinCFlow.jl) (Pseudo-inCompressible Flow solver), developed for conducting idealized atmospheric simulations, integrates the Boussinesq, pseudo-incompressible, and compressible equations using a finite volume approach and a semi-implicit time-stepping scheme. Due to MPI parallelization in all spatial dimensions, high-resolution, wave-resolving simulations can be conducted efficiently. The use of a terrain-following coordinate system allows for arbitrary surface topography.
A unique feature of [PinCFlow.jl](https://github.com/Atmospheric-Dynamics-GUF/PinCFlow.jl) is its numerical realization of multiple-scale WKB theory for the parameterization of gravity-wave impacts on mean-flow dynamics using MS-GWaM (Multi-Scale Gravity-Wave Model). Here, gravity waves are approximated using a ray-tracing technique, where so-called ray volumes propagate through a six-dimensional phase-space. Unlike conventional single-column, steady-state parameterizations, MS-GWaM accounts for wave transience and horizontal propagation.
Current research and development focuses on the generation of orographic gravity waves, the interaction between gravity waves and turbulence, and their impact on ice cloud formation and tracer transport.
Originally developed in Fortran 90, the solver has been fully translated and optimized as a Julia package. Now officially registered and accompanied by extensive documentation, [PinCFlow.jl](https://github.com/Atmospheric-Dynamics-GUF/PinCFlow.jl) provides a modern, high-performance framework for research in computational fluid dynamics. Idealized Atmospheric Flow and Gravity-Wave Modeling with PinCFlow.jl | Steiger](https://i.ytimg.com/vi/h6SKj25kL6w/mqdefault.jpg)









