Your image description


LambdaSim's main goal is to act as a digital twin of the world from a macroeconomic perspective and to simulate future scenarios by taking advantage of agent based modeing and various methods from multivariat data analysis.


Throughout human history, we have made remarkable progress in solving analytical problems. For example, in 1977, we successfully calculated the precise trajectories for Voyager 1 and 2, ensuring their successful journeys through space. Similarly, we have built increasingly more advanced architecture thanks to the continual progress of material science, CAD software and general physics. However, while these endeavors are intricate and require detailed calculations, they aren't truly complex.

-->

Here is a peek at what's driving this project along with a project timeline.

Complexity arises when there are nonlinear relationships, feedback loops, and unstable dynamics. Examples range from predicting traffic jams in a bustling city to understanding how the weather will unfold based solely on initial conditions.


Fortunately, we can make significant progress in predicting complex systems by breaking down the system into smaller components often called agents. These agents follow a simpler set of rules regarding how they act with their environment. This approach allows us to iteratively simulate the behavior of the entire system. By trying different values for the variables that we can change in real-world scenarios within the simulation, we gain valuable insights about how the system as a whole will be affected. As a consequence, this insight enables us to make better, more informed decisions in the real world.


In science, testing a hypothesis is crucial for proving or disproving a theory. However, in economics, especially on large scales, it is both analytically unpredictable and often very costly or impossible to test new government policies, such as those listed here, before analyzing their real-world implementation results and aftereffects. Additionally, in democratic governments, where new politicians are elected every four years, a policy might be reversed before any meaningful impact, positive or negative, can be measured.


This is the problem that Lambdasim aims to tackle. By creating detailed and dynamic models of economic systems, simulations can predict the potential outcomes of various policies without the need for real-world implementation. These simulations can account for numerous variables and their interactions, providing a more comprehensive analysis of possible results. This approach allows policymakers to evaluate the potential impacts of their decisions in a controlled environment, reducing the risks, costs, and ambiguities associated with trial-and-error in real-world policy testing.

Highly relevant paper

The paper applies an Agent Based Model (ABM) model to simulate Austria’s medium-run economic recovery from COVID-19 lockdowns, finding the ABM’s results consistent with official forecasts by Austrian institutions.

Key Takeaways

Why It’s Important

Read the Full Paper

Download the PDF of "Economic Forecasting with an Agent-Based Model"



Check out LambdaSim on GitHub

Project Timeline

(Click the blue boxes to expand)