Risk-Adjusted Excess Return Analysis and Robust Evaluation of Four-Asset Momentum Rotation Strategy

Based on the backtest results you provided and existing research, I present conclusions on whether the strategy has risk-adjusted excess returns, clarify its applicability and limitations, and offer recommendations for robust evaluation.

1. Conclusions on Whether There Are Risk-Adjusted Excess Returns

• Backtest Perspective: During the sample period you mentioned (from around 2013/2014 to present), this four-asset momentum rotation strategy (annualized return 24.24%, Sharpe ratio 1.08) indeed showed higher returns and higher risk-adjusted returns (higher Sharpe ratio) compared to the equal-weighted portfolio (annualized return 11.63%, Sharpe ratio 0.94). If measured together with indicators like maximum drawdown and volatility, it is necessary to confirm whether drawdown and volatility have improved simultaneously. However, based solely on the numbers you provided, there is obvious risk-adjusted excess within the sample.
• External Evidence: Asset-level momentum (cross-asset momentum) is generally considered to have long-term anomalies in academia and industry (e.g., Moskowitz et al.'s research on time-series momentum; cross-border and cross-asset momentum also have empirical support in different markets), but the magnitude and persistence are affected by cycles, costs, and institutions. Therefore, “there is certain excess in the long run” is a relative consensus in academic literature, but the excess magnitude and stability vary greatly.
• Key Constraints: Historical results do not equal the future. The excellent performance of backtesting may be determined by the specificity of the sample period, parameter and rule selection, transaction cost and slippage assumptions, rebalancing frequency and threshold settings, etc. Whether it will recur in the future requires stricter forward-looking testing and continuous tracking.

2. Applicability Evaluation (When It Is More Likely to Be Effective)

• Market Conditions:
  • Trend and Volatility Environment: Momentum signals are more effective when there are relatively sustained and identifiable trend segments in macro and asset prices, or when volatility rises leading to increased value of allocation switching; fast mean reversion or frequent switching environments will reduce effectiveness.
  • Liquidity and Cost: Target ETFs with sufficient liquidity, controllable bid-ask spreads and impact costs, and moderate transaction frequency (e.g., monthly or longer) are more suitable for low-cost execution.
  • Correlation and Diversification: The correlation structure between the four assets is relatively stable, and it can provide diversification in terms of cycle, defense, growth, etc.; rotation will have a better chance of bringing “better risk-return position switching”.
• Underlying Assets and Rules:
  • Investability and Data Quality of Underlying Assets: Cross-asset ETFs from large public funds such as Guotai Fund, Huatai-PineBridge Fund, Huaan Fund, and Pengyang Fund usually have advantages in tracking error and liquidity, but it is still necessary to verify the historical continuity and scale stability of each underlying asset.
  • Signals and Rebalancing:
    • Focusing on return momentum (e.g., relative strength in the past N months) and superimposing risk or volatility reduction helps improve robustness;
    • Avoid excessive complexity (high parameters, overfitting) and overly short rebalancing cycles (costs erode returns).
• Portfolio and Account Constraints:
  • Position and Leverage: As you mentioned, moderate leverage can amplify returns, but it will also amplify drawdowns and tail risks; risk tolerance, financing costs, and forced liquidation constraints need to be strictly evaluated.
  • Taxation and Regulation: Stamp duty, dividend tax, capital gains treatment, pledge and leverage rules for domestic institutional and individual accounts will affect the strategy’s net value and executability.

3. Limitations and Risks (Points to Be Alerted)

• Overfitting and Sample Specificity: The strategy’s performance may be over-adapted to historical data (the “optimal version” selected from multiple parameters, multiple thresholds, and multiple alternative underlying assets). Excellent performance within the sample does not mean robustness outside the sample.
• Transaction Costs and Slippage: If backtesting does not fully include commissions, bid-ask spreads, and market impact, especially when switching frequently or using high leverage, the actual return may be significantly lower than the backtest.
• Institutional and Structural Changes: Listing rules, ETF product iterations, changes in fees and liquidity, and adjustments to cross-border and derivative policies will change the strategy’s operating environment.
• Behavior and Crowding: If similar strategies are widely used, signal crowding will weaken their advantages and cause resonance and踩踏 (panic selling) in extreme market conditions.
• Single Risk Exposure: If the momentum strategy has systematic exposure to certain common factors between assets (e.g., growth/value, term/credit, macro risks), it may suffer synchronous losses in tail events.

4. How to Conduct Robust Evaluation (Practical Recommendations)

• Out-of-Sample and Time-Segmented Testing:
  • Reserve at least 1/3 of the time period for out-of-sample testing, or use rolling windows (e.g., re-evaluate parameter robustness every 3-5 years) to observe the performance of parameter extrapolation.
  • Divide cycles (bull market/bear market/sideways market, interest rate hike/rate cut cycle, etc.) to see the consistency of the strategy’s performance in different sub-stages.
• Cost and Slippage Stress Testing:
  • Introduce relatively conservative commission, bid-ask spread, and market impact models in backtesting (especially during periods of low trading volume or large scale).
  • Sensitivity Analysis: Assume increased transaction frequency and increased impact, and observe changes in the strategy’s net return and Sharpe ratio.
• Robustness of Parameters and Rules:
  • Parameter Scanning: Conduct a wide range of grid searches for observation period length, rebalancing frequency, thresholds, etc., to find “flat intervals” (not overly sensitive to parameters);
  • Single Signal vs. Multiple Signals: Try adding volatility filtering, trend confirmation, or risk parity dimensions to see if it improves robustness rather than just fitting history.
• Risk Indicators and Stress Scenarios:
  • Monitor maximum drawdown, downside capture, conditional Sharpe ratio, VaR/ES, etc., and design drawdown control mechanisms (e.g., position limits, volatility targets).
  • Stress Scenario Testing: Portfolio profit and loss and leverage risks under extreme interest rate changes, credit events, liquidity tightening, and other scenarios.
• Operation and Execution:
  • Clarify rebalancing discipline and execution windows (avoid chasing ups and downs intraday to increase impact).
  • Set strict risk limits and liquidation lines for the leverage part, and conduct liquidity stress tests.
  • Continuously monitor changes in underlying asset fees, scale, and liquidity, and replace or adjust underlying assets if necessary.

5. Practical Conclusions

• During the sample period you provided, the momentum rotation strategy showed higher annualized returns and higher Sharpe ratios compared to the equal-weighted allocation, and historical evidence supports its potential to have risk-adjusted returns.
• However, this advantage is highly dependent on market environment, execution costs, and parameter robustness. If used for long-term portfolio management, it should:
  • Establish a strict out-of-sample and time-segmented testing framework;
  • Introduce conservative cost and slippage assumptions and conduct stress tests;
  • Control leverage and drawdowns to avoid excessive concentration of a single strategy;
  • Regularly review and update rules to avoid strategy aging and overfitting.
• If used for accounts with large capital or strict regulatory constraints, it is recommended to adopt a progressive pilot, small-scale verification, and continuous tracking approach, combined with macro and factor perspectives for dynamic adjustment.

6. References (Based on Your Provided Context and Public Information)

• Jinling API Data [0]: Contains historical and real-time data on asset prices, volatility, and correlations in the A-share market, used for strategy backtesting and risk assessment.
• Backtest analysis post you provided (in-sample results of four-asset rotation strategy: annualized 24.24%, Sharpe ratio 1.08 vs. equal-weighted 11.63%, Sharpe ratio 0.94): Constitutes direct historical performance basis.
• Moskowitz, T. J., Ooi, Y. H., & Pedersen, L. H. (2012). Time series momentum. Journal of Financial Economics, 104(2), 228-250. (Academic literature on time-series momentum).
• Public materials from various fund companies (Guotai Fund, Huatai-PineBridge Fund, Huaan Fund, Pengyang Fund) and relevant ETF product descriptions, fee and liquidity disclosures: Used for underlying asset investability and cost assessment.

Note: Regarding the specific backtest details in this post (ETF used, rebalancing frequency, cost assumptions, leverage settings, sample start and end points, and time-segmented performance), please refer to the author’s complete disclosure; the above recommendations are based on general best practices to enhance the robustness of strategy evaluation and execution, but do not constitute investment advice. Future returns are uncertain, and past performance does not indicate future results.