Paper: arXiv 2410.14841
Authors: Yizhan Shu, John M. Mulvey
Abstract
This article explores dynamic factor allocation by analyzing the cyclical performance of factors through regime analysis. The authors focus on a U.S. equity investment universe comprising seven long-only indices representing the market and six style factors: value, size, momentum, quality, low volatility, and growth. Their approach integrates factor-specific regime inferences of each factor index’s active performance relative to the market into the Black-Litterman model to construct a fully-invested, long-only multi-factor portfolio. First, the authors apply the sparse jump model (SJM) to identify bull and bear market regimes for individual factors, using a feature set based on risk and return measures from historical factor active returns, as well as variables reflecting the broader market environment. The regimes identified by the SJM exhibit enhanced stability and interpretability compared to traditional methods. A hypothetical single-factor long-short strategy is then used to assess these regime inferences and fine-tune hyperparameters, resulting in a positive Sharpe ratio of this strategy across all factors with low correlation among them. These regime inferences are then incorporated into the Black-Litterman framework to dynamically adjust allocations among the seven indices, with an equally weighted (EW) portfolio serving as the benchmark. Empirical results show that the constructed multi-factor portfolio significantly improves the information ratio (IR) relative to the market, raising it from just 0.05 for the EW benchmark to approximately 0.4. When measured relative to the EW benchmark itself, the dynamic allocation achieves an IR of around 0.4 to 0.5. The strategy also enhances absolute portfolio performance across key metrics such as the Sharpe ratio and maximum drawdown.
Complexity vs Empirical Score
- Math Complexity: 7.5/10
- Empirical Rigor: 8.0/10
- Quadrant: Holy Grail — high math complexity, high empirical rigor
Why this score: The paper employs advanced statistical methods like the Sparse Jump Model and Black-Litterman with feature weighting, requiring sophisticated mathematical derivations. It is heavily data-driven, with detailed empirical backtesting on real indices, incorporating transaction costs, and reporting specific performance metrics like Sharpe ratios and Information Ratios.
Research Flowchart
flowchart TD A["<b>Research Goal</b><br>Dynamic Factor Allocation<br>via Regime-Switching"] --> B["<b>Data Inputs</b><br>U.S. Equity Universe<br>7 Long-Only Indices<br>(Market + 6 Style Factors)"] B --> C["<b>Regime Identification</b><br>Sparse Jump Model (SJM)<br>Identifies Bull/Bear Regimes<br>for Each Factor"] C --> D["<b>Hyperparameter Tuning</b><br>Hypothetical Single-Factor<br>Long-Short Strategy<br>Validates Regimes"] D --> E["<b>Dynamic Allocation</b><br>Black-Litterman Framework<br>Integrates Regime Signals<br>& Market Views"] E --> F["<b>Key Outcomes</b><br>IR: 0.05 → 0.4 (vs Market)<br>IR: 0.4-0.5 (vs EW Benchmark)<br>Improved Sharpe & Drawdown<br>Low Factor Correlation"]
Evidence Card
What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card
Proposes
The paper proposes a dynamic factor allocation strategy that uses sparse jump model regime inferences integrated into a Black‑Litterman framework to adjust long‑only multi‑factor weights.
Source sentence
Motivated by the cyclical behavior of factor returns, this article proposes a dynamic factor allocation strategy through regime analysis based on each factor’s active performance, synthesizing these factor-specific regime inferences using the Black‑Litterman model.
Assumes 4
- Regime inference can be applied with a one‑day delay (T+2) and is assumed to persist in the near future. model
Source sentence
we assume that this inferred regime sT can only be applied on day T + 2 with a one‑day delay – meaning that we estimate sT at the end of day T and rebalance the portfolio accordingly at the end of the next day, T + 1.
- The portfolio is long‑only, fully invested across the market and six factor indices. execution
Source sentence
Our dynamic strategy maintains a long‑only, fully‑invested allocation among the market and six factor indices.
- Features representing market environment are available in real time without delay. data
Source sentence
These variables are based on trading markets, updated daily, and are readily accessible in real time without delay.
- Transaction costs are a flat 5 basis points per side. execution
Source sentence
A transaction cost of 5 basis points is applied for both buying and selling.
Data
2007 – 2024 · U.S. equities · equities
- Seven long‑only indices (market + six style factors) market-prices public
Source sentence
Our empirical study focuses on a U.S. equity investment universe comprising seven long‑only indices representing the market and six style factors – value, size, momentum, quality, low volatility, and growth.
- FRED macro‑economic variables macro public
Source sentence
We use the aforementioned variables, easily sourced from the FRED Database, as proxies for the general market environment.
FRED Bloomberg
Costs
Modeled
5 basis points per side applied to both buying and selling.
Source sentence
A transaction cost of 5 basis points is applied for both buying and selling.
Finds 3
Sharpe 0.65 · annual return 12.5% · max drawdown 50.5%
- The dynamic multi‑factor portfolio raises the information ratio from about 0.05 (EW benchmark) to roughly 0.4‑0.5. performance
Source sentence
Empirical results show that the constructed multi‑factor portfolio significantly improves the information ratio (IR) relative to the market, raising it from just 0.05 for the EW benchmark to approximately 0.4. When measured relative to the EW benchmark itself, the dynamic allocation achieves an IR of around 0.4 to 0.5.
- Positive Sharpe ratios are achieved for all single‑factor long‑short strategies with low inter‑factor correlations. performance
Source sentence
We achieve positive Sharpe ratios across all factors, with low correlations among them, demonstrating the high accuracy of our regime inferences.
- Dynamic allocation improves absolute performance metrics such as Sharpe ratio and reduces maximum drawdown compared to a static benchmark. performance
Source sentence
Empirical results show that our dynamic allocation strategy outperforms a static approach, enhancing key metrics such as the information ratio and Sharpe ratio while reducing maximum drawdown.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 17 related papers). Coverage is partial, so this is neither replication nor refutation.
Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.