Towards robust statistical inference for complex computer models
Abstract
Ecologists increasingly rely on complex computer simulations to forecast ecological systems. To make such forecasts precise, uncertainties in model parameters and structure must be reduced and correctly propagated to model outputs. Naively using standard statistical techniques for this task, however, can lead to bias and underestimation of uncertainties in parameters and predictions. Here, we explain why these problems occur and propose a framework for robust inference with complex computer simulations. After having identified that model error is more consequential in complex computer simulations, due to their more pronounced nonlinearity and interconnectedness, we discuss as possible solutions data rebalancing and adding bias corrections on model outputs or processes during or after the calibration procedure. We illustrate the methods in a case study, using a dynamic vegetation model. We conclude that developing better methods for robust inference of complex computer simulations is vital for generating reliable predictions of ecosystem responses.
What the paper shows and why it matters (AI-generated)
Complex ecological simulations amplify structural model error in ways that make naive statistical calibration confidently wrong — biased parameters, underestimated uncertainty — and this paper works out exactly why, then proposes two concrete fixes: rebalancing unevenly sized data and correcting for bias in model outputs or processes, demonstrated on a dynamic vegetation model. The framework has since become a reference point for the authors' own later work on calibration and uncertainty quantification, including a 2026 review of the whole field it helped set up.