Idiosyncratic Market Effects in Global Time Series Momentum

Paper: arXiv 2611.15646

Authors: Andrew Butler, Maciej Zawadzki, Adam Butler

Abstract

Time series momentum is a pervasive explanatory factor in financial markets whereby a market’s recent historical average return positively predicts future returns. Recent applications of statistical learning augment the traditional linear model definition with a more descriptive global nonlinear factor model. We extend the work of \citet{MoskowitzNonlinear} and demonstrate that models fit at the individual market level exhibit response curves that are distinct from the global factor model. We highlight the heterogeneity of forward market returns as a function of time series momentum and provide a framework that enables each feature-market pair to express its unique prediction behaviour. Empirical results demonstrate that, despite substantially smaller sample sizes, fitting nonlinear models to individual markets is a robust and often overlooked modelling approach that can result in a marginal improvement in out-of-sample performance.

Complexity vs Empirical Score

• Math Complexity: 6.0/10
• Empirical Rigor: 7.0/10
• Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper presents a clear extension of existing work, introducing a novel perspective on individual market modeling for time series momentum. It demonstrates solid empirical methodology with a focus on out-of-sample performance, though the statistical significance of the marginal improvement is noted as small.

Research Flowchart

  flowchart TD
    A[Research Goal: Investigate Idiosyncratic Market Effects in Time Series Momentum] --> B{Methodology: Individual Market Nonlinear Models};
    B --> C[Data/Inputs: Global Time Series Momentum Data (Futures / Multi-Asset)];
    C --> D{Computational Process: Fit Nonlinear Models to Individual Markets & Global Factor Model};
    D --> E[Outcome 1: Heterogeneity of Response Curves (Individual vs. Global)];
    D --> F[Outcome 2: Improved Out-of-Sample Performance with Individual Market Models];
    E & F --> G[Key Finding: Individual Market Models are a Robust, Overlooked Approach];

Cited References (381)

1. Nonlinear Time Series Momentum (2025)
2. Loss Harvesting or Gain Deferral? A Surprising Source of Tax Benefits of Tax-Aware Long-Short Strategies (2024)
3. Tax-Aware Dynamic Asset Allocation (2016) — link
4. Portfolio construction as linearly constrained separable optimization (2023) — link
5. Tax-Aware Portfolio Construction via Convex Optimization (2021) — link
6. Tax-Aware Portfolio Construction: A Multi-Asset Approach (2025)
7. Latent Period (reaction Time) (2009)
8. Hybrid Intelligent Systems for Stock Market Analysis (2001)

0 of 381 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2570 linked papers.