Instrumental variables uncover causal effects in continuous time processes

Instrumental Variables in Continuous-Time Stochastic Processes

Machine Learning

Summary

Understanding cause and effect is hard when things change continuously over time and hidden factors affect the outcomes. The authors developed a new mathematical way to find causes in such continuous-time systems using special tools called instrumental variables, which act like clues to untangle hidden influences. Their method works by examining the tiny changes that happen at every instant and can find cause-effect links even when traditional methods, which look only at snapshots in time, get confused. They tested their approach on specific models and showed it can correctly find causal effects where other common techniques fail.

What this means in practice

  • For quantitative finance teams: Identify causal relationships in financial models that change continuously and have hidden confounders using the new continuous-time instrumental variable framework.
  • For healthcare data analysts: Estimate causal effects in patient monitoring systems where physiological signals evolve continuously and confounding variables are present.

Authors

Lars Pennig, Søren Wengel Mogensen, Niki Kilbertus

Abstract

Causal questions often arise in systems that evolve in continuous time. While instrumental variable (IV) methods provide a principled approach to identify causal effects in static or discrete-time settings, even in the presence of unobserved confounding, they do not yet extend to continuous-time dynamics. In this work, we develop an IV framework for stationary continuous-time stochastic processes parameterized by stochastic differential equations (SDE), where the instrument itself takes the form of a stochastic process. We encode the structural sparsity and causal effects directly at the infinitesimal level of the drift, and contrast these assumptions with the IV assumptions on the discretized version of the same process. Based on these structural assumptions, we derive moment restrictions using path-valued instruments. We show how these restrictions simplify for Ornstein-Uhlenbeck processes and present an identification condition together with a corresponding consistent estimation method. Finally, we extend our estimation framework to nonlinear causal effects by expressing the path-based moment restrictions leveraging the signature kernel. Empirically, our method recovers causal effects in confounded continuous-time systems where maximum-likelihood-based estimation and discrete-time IV methods are systematically biased.