Skip to main content

Causal Inference

Causal inference is the process of determining the causal effect of one variable on another (i.e. how much one variable causes another). This is different from the standard approach of just seeing if two variables are correlated. It's important because most business questions are causal: they're about what action/decision/intervention you want to take, optimisations, or prediction in a changing world, or just understanding why something has happened. For all of this only causal methods will give you the answer you want.

Causal inference is a fundamental problem in statistics and machine learning, and it is essential for making decisions based on data. However, much of the existing tooling for causal inference is highly technical and difficult to understand for anyone who is not a causal AI expert. Furthermore, many of the existing tools are focused on inferring causal relationships rather than estimating the outcomes of actions, which is critical for decision-making and AI that can reason.

CausaDB is designed to make causal inference accessible to everyone, by providing a simple and intuitive interface for estimating causal effects and making data-driven decisions, without the typical knowledge burden that comes with causal AI. It also goes beyond the limitations of traditional causal inference methods by allowing users to simulate the effects of actions, optimise decisions based on causal models, and explain how AI models make decisions.