Forecastability as an Information-Theoretic Limit on Prediction

TL;DR

This paper establishes an information-theoretic framework for understanding the limits of forecastability based on the mutual information between future observations and available information, independent of specific models.

Contribution

It derives exact properties of forecastability profiles, linking them to process structure and providing a diagnostic for method adequacy and information sufficiency.

Findings

Forecastability profile reflects dependence structure and can be non-monotonic.

Compression of information reduces forecastability.

Near-zero forecastability implies no method can outperform the baseline.

Abstract

Forecasting is usually framed as a problem of model choice. This paper starts earlier, asking how much predictive information is available at each horizon. Under logarithmic loss, the answer is exact: the mutual information between the future observation and the declared information set equals the maximum achievable reduction in expected loss. This paper develops the consequences of that identity. Forecastability, defined as this mutual information evaluated across horizons, forms a profile whose shape reflects the dependence structure of the process and need not be monotone. Three structural properties are derived: compression of the information set can only reduce forecastability; the gap between the profile under a finite lag window and the full history gives an exact truncation error budget; and for processes with periodic dependence, the profile inherits the periodicity. Predictive loss decomposes into an irreducible component fixed by the information structure and an approximation component attributable to the method; their ratio defines the exploitation ratio, a normalised diagnostic for method adequacy. The exact equality is specific to log loss, but when forecastability is near zero, classical inequalities imply that no method under any loss can materially improve on the unconditional baseline. The framework provides a theoretical foundation for assessing, prior to any modelling, whether the declared information set contains sufficient predictive information at the horizon of interest.