Lepidoptera enables causal simulation with mixed data types
lepidoptera: A Domain-Agnostic Engine for Causal Simulation with Mixed Variable Types, Vine Copula Dependence, and Individual-Level Counterfactual Propagation
Artificial IntelligenceMachine Learning
Summary
Simulating cause-and-effect relationships can be tricky when data comes in many forms, like numbers or categories. The authors created lepidoptera, a software tool that handles different types of data together while following cause-and-effect rules. It uses smart math to understand complex dependencies and lets users ask what-if questions about individual cases without extra assumptions. The tool works in both R and Python and can be combined with other models for various applications like medical study planning and emergency management.
What this means in practice
- •For clinical trial planners: Simulate outcomes under interventions combining numeric and categorical patient data to optimize trial design and estimate treatment effects at the individual level.
- •For hospital capacity managers: Model downstream effects of changes in emergency ward resources by simulating causal relationships across diverse data types affecting patient flow.
- •For insurance risk analysts: Create detailed simulations integrating different data types to assess individual risk scenarios and support underwriting decisions using causal inference.$Commercial implications: Enables sale of advanced risk assessment tools tailoring insurance offers based on individual counterfactual analyses.
Authors
Danilo Augusto Sarti
Abstract
lepidoptera is an open-source R and Python package for simulation from structural causal models with heterogeneous variable types. The package is built around a typed mechanism architecture, which separates the causal graph from the generating rule at each node and exposes each rule as an inspectable, serialisable object. A pre-built mechanism library covers all nine combinations of continuous, categorical, and ordinal parent and child types, removing the need for users to implement type-specific sampling from scratch. A vine copula mechanism estimates the full joint dependence structure from observed data using pair-copula constructions and the Rosenblatt conditional-simulation algorithm, supporting over twenty copula families. Individual-level counterfactual queries are supported through partial propagation, which freezes the random draws of unaffected nodes and replays only the downstream consequences of an intervention, thereby separating the individual treatment effect (ITE) estimand from the population-level average treatment effect (ATE) without additional model assumptions. An extension layer allows any pre-fitted predictive model to serve as a generative mechanism via a unified callable interface. Three worked illustrations spanning clinical trial planning, emergency-ward capacity analysis, and epidemiological mediation analysis demonstrate the package's scope.