Nonlinear reconciliation: Error reduction theorems

TL;DR

This paper develops formal theorems demonstrating error reduction in nonlinear forecast reconciliation, extending previous linear results to complex nonlinear constraints and providing a practical Python package for implementation.

Contribution

It introduces new error reduction theorems for nonlinear reconciliation on various hypersurfaces and manifolds, filling a key theoretical gap in probabilistic forecast reconciliation.

Findings

Error reduction theorems for hypersurfaces with constant-sign curvature

Error reduction results for hypersurfaces with non-constant-sign curvature

Implementation of theorems in the JAX-based Python package JNLR

Abstract

Forecast reconciliation, an ex-post technique applied to forecasts that must satisfy constraints, has been a prominent topic in the forecasting literature over the past two decades. Recently, several efforts have sought to extend reconciliation methods to the probabilistic settings. Nevertheless, formal theorems demonstrating error reduction in nonlinear constraints, analogous to those presented in Panagiotelis et al.(2021), are still lacking. This paper addresses that gap by establishing such theorems for various classes of nonlinear hypersurfaces and vector-valued functions. Specifically, we derive an exact analog of Theorem 3.1 from Panagiotelis et al.(2021) for hypersurfaces with constant-sign curvature. Additionally, we provide an error reduction theorem for the broader case of hypersurfaces with non-constant-sign curvature and for general manifolds with codimension > 1. To support reproducibility and practical adoption, we release a JAX-based Python package, JNLR, implementing the presented theorems and reconciliation procedures.