Global Seasonality Factor Investing

Paper: arXiv 2611.15657

Authors: Andrew Butler, Adam Butler, Maciej Zawadzki

Abstract

We study seasonality as a factor for predicting risk-adjusted returns across global futures markets. We document five seasonality anomalies that, in aggregate, are pervasive across bond, commodity and equity markets. We demonstrate that each seasonality factor produces excess returns that are not reasonably explained by established global time series carry, momentum and skewness factors. We apply recent innovations in statistical learning and fit nonlinear seasonality models at the individual market level, thereby allowing each feature-market pair to express its unique seasonal prediction pattern. We discuss candidate economic rationales for the existence and persistence of each seasonality anomaly. Walk-forward out-of-sample testing demonstrates that fast-moving seasonal patterns, such as day-of-week and day-of-month, produce strong gross performance but with extremely high turnover, resulting in negative performance net of transaction costs. In contrast, slower-moving seasonal patterns, such as day-of-quarter, day-of-half-year and day-of-year, produce positive gross performance and, with the exception of day-of-quarter, positive net performance. An equal-weight ensemble of seasonality models produces an out-of-sample Sharpe ratio that is competitive with, and uncorrelated to, established global time series factor portfolios. Empirical results demonstrate that adding seasonality to a global factor portfolio increases the out-of-sample Sharpe ratio by approximately $30%$ over the full evaluation period.

Complexity vs Empirical Score

• Math Complexity: 6.5/10
• Empirical Rigor: 8.0/10
• Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper presents a robust empirical study of seasonality factors across global futures markets, utilizing advanced statistical learning techniques. Its novelty lies in consolidating various seasonal anomalies into a unified framework and demonstrating their uncorrelated nature with established factors. The methodology is clearly articulated, supporting strong reproducibility and empirical rigor.

Research Flowchart

  flowchart TD
    A[Research Goal: Predict Risk-Adjusted Returns using Seasonality in Global Futures] --> B{Methodology: Identify & Test Seasonality Anomalies};
    B --> C[Data/Inputs: Global Futures Markets];
    C --> D{Computational Process: Statistical Learning for Nonlinear Seasonality Models};
    D --> E[Outcome 1: Document 5 Pervasive Seasonality Anomalies];
    E --> F[Outcome 2: Excess Returns Unexplained by Established Factors];
    F --> G[Outcome 3: Slower Seasonality Patterns Lead to Positive Net Performance];
    G --> H[Outcome 4: Seasonality Ensemble Improves Sharpe Ratio of Global Factor Portfolio by ~30%];

Cited References (398)

1. Seasonality and the valuation of commodity options (2013) — link
2. Stock Trader’s Almanac 2021 (2020)
3. Seasonality in government bond returns and factor premia (2017) — link
4. Return Seasonalities (2016)
5. Seasonality in the cross-section of stock returns (2008)
6. The Halloween Indicator, Sell in May and Go Away: Another Puzzle (2002)
7. Stock Returns and the Weekend Effect (1980)
8. Seasonality and momentum across national equity markets (2022)

1 of 398 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2570 linked papers.