---
title: "The Spectrum Is Not Enough: When Context Helps Time-Series Forecasting"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2607.13006v1"
permalink: "https://feed.astralor.com/items/itm_beeb6a2bd85d4f46b2ebb127"
published: "2026-07-14T17:50:28.000Z"
score: 76
channel: "ai"
tags: ["Foundation Models","Time Series Forecasting","Retrieval-Augmented","Phase Randomization","Coverage Deficit","Spectral Analysis"]
---

# The Spectrum Is Not Enough: When Context Helps Time-Series Forecasting

## AI commentary

> This paper timely corrects a common misuse of spectral indices in the industry. For engineers and researchers working on time-series forecasting, understanding that context value cannot be inferred from spectrum alone is critical; a more nuanced diagnostic is needed. The coverage deficit could become a practical tool, especially for rapid assessment before deploying retrieval-augmented or foundation models.

## At a glance

This paper argues that common spectral indices cannot measure the value of context (e.g., longer lookback, retrieval plug-in, pretrained models) for time-series forecasting, because context value depends on the operating point, not the series itself. By constructing surrogate pairs via phase randomization that preserve the spectrum, the authors show that spectral indices are invariant to higher-order structure. They propose a label-free diagnostic, the coverage deficit, which measures beyond-spectrum structure as the gain of analog over linear prediction. On seven benchmarks, window-keyed retrieval's value collapses across surrogates (ECL median from +33% to -35%, p<10^{-40}) while all spectral indices remain frozen; a foundation model's value partially survives; the structure term predicts the sign of beyond-spectrum value. The paper contributes a distinction, a controlled comparison, and a deployment diagnostic without introducing new forecasters.

## Key points

- Spectral indices cannot answer whether adding context helps, because context value depends on the operating point.
- Phase randomization preserves the spectrum but destroys higher-order structure; spectral indices remain unchanged while retrieval/foundation model values collapse.
- The paper proposes coverage deficit, a label-free diagnostic measuring beyond-spectrum structure as the gain of analog over linear prediction.
- Experiments show that window-keyed retrieval's value drops from +33% to -35% across surrogates, while all spectral indices freeze; foundation model values partially survive.

## Full text

A growing family of indices scores how predictable a series is from its spectrum. Practitioners increasingly read these scores as answering a different question: whether \emph{adding context}, a longer lookback, a retrieval plug-in, or a pretrained model, will help. These are not the same question. The value of context is a property of the operating point, not of the series. Any index built from the power spectrum is invariant under phase randomization, whereas the beyond-second-order value that retrieval and foundation models supply is not, because a phase-randomized series is asymptotically Gaussian. We state this as an impossibility result and isolate it with surrogate pairs that fix the spectrum and the marginal by construction. We then give a label-free, configuration-level diagnostic, the coverage deficit, whose principal term measures beyond-spectrum structure as the gain of analog over linear prediction. On seven benchmarks the prediction holds: window-keyed retrieval's value collapses across surrogate pairs (ECL median $+33\%\!\to\!-35\%$, $p{<}10^{-40}$) while every spectral index stays frozen; a foundation model's value splits into a surviving second-order part and a small beyond-linear margin that collapses; a longer linear window's value survives. Leave-one-dataset-out, the structure term predicts the sign of beyond-spectrum value where the spectral indices trail it, and the reverse holds for the second-order mechanism. We introduce no new forecaster; the contribution is the distinction, a controlled comparison, and a diagnostic for the deployment decision. Code: https://anonymous.4open.science/r/SINE.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_beeb6a2bd85d4f46b2ebb127