Laplace’s Demon and the Adventures of His Apprentices

Abstract

The sensitive dependence on initial conditions (SDIC) associated with nonlinear models imposes limitations on the models' predictive power. We draw attention to an additional limitation than has been under-appreciated, namely structural model error (SME). A model has SME if the model-dynamics differ from the dynamics in the target system. If a nonlinear model has only the slightest SME, then its ability to generate decision-relevant predictions is compromised. Given a perfect model, we can take the effects of SDIC into account by substituting probabilistic predictions for point predictions. This route is foreclosed in the case of SME, which puts us in a worse epistemic situation than SDIC.

Scotland׳s hydropower: Current capacity, future potential and the possible impacts of climate change

“…This is a very common problem for climate impact assessments in general, and the development of robust methods for assessing uncertainties within complex modelling chains is an area of active research e.g. [47,60].…”

When are Purely Predictive Models Best?

“…It has long been thought a chaotic system, in other words that outcomes are indefinitely sensitive to exact initial conditions (Lorenz 1969). More recently, it has also been convincingly argued that weather predictions are (often) also indefinitely sensitive to model errors-that is, even tiny inaccuracies in a model can lead to very large errors in the predictions made by that model (Frigg et al 2014). These are obviously major challenges for prediction.…”

Expert Judgment for Climate Change Adaptation

“…The alternative to SEE is not objective calculation; the alternatives are unstructured expert judgment (as in the IPCC report) or unstructured nonexpert judgment (when model-world results are communicated to policymakers ill-equipped to assess their real-world importance). Reliable objective calculation of probabilities based on nonlinear model output has been argued elsewhere to be infeasible (Frigg et al 2014). SEE does not guarantee trustworthiness, but it can compensate for (some of ) the (known) inadequacies of climate models.…”