The Future Has Thicker Tails than the Past: Model Error As Branching Counterfactuals

TL;DR

This paper explores how recursive epistemic uncertainty in forecasting leads to fat-tailed distributions, significantly impacting risk assessment and the reliability of traditional probabilistic methods.

Contribution

It introduces a framework for understanding how nested counterfactual errors produce fat tails, challenging conventional assumptions about distribution shapes in forecasting.

Findings

Nested error recursions lead to fat-tailed distributions.

Even minimal branching error rates cause explosive higher moments.

Ignoring regress in uncertainty underestimates small probabilities.

Abstract

Ex ante forecast outcomes should be interpreted as counterfactuals (potential histories), with errors as the spread between outcomes. Reapplying measurements of uncertainty about the estimation errors of the estimation errors of an estimation leads to branching counterfactuals. Such recursions of epistemic uncertainty have markedly different distributial properties from conventional sampling error. Nested counterfactuals of error rates invariably lead to fat tails, regardless of the probability distribution used, and to powerlaws under some conditions. A mere .01% branching error rate about the STD (itself an error rate), and .01% branching error rate about that error rate, etc. (recursing all the way) results in explosive (and infinite) higher moments than 1. Missing any degree of regress leads to the underestimation of small probabilities and concave payoffs (a standard example of which is Fukushima). The paper states the conditions under which higher order rates of uncertainty (expressed in spreads of counterfactuals) alters the shapes the of final distribution and shows which a priori beliefs about conterfactuals are needed to accept the reliability of conventional probabilistic methods (thin tails or mildly fat tails).