- The report introduces three quantitative models: Commodity Term Structure, Commodity Time-Series Momentum, and Commodity Cross-Sectional Inventory. These models are combined into a Composite Commodity Strategy using equal weighting of the three sub-strategies[25][26][28] - **Commodity Term Structure Model**: This model is constructed based on the roll yield factor to capture the contango and backwardation states of commodities. It dynamically goes long on commodities with high roll yields and short on those with low roll yields[26][30][33] - **Commodity Time-Series Momentum Model**: This model uses multiple technical indicators to capture medium- and long-term trends in domestic commodities. It dynamically goes long on assets with upward trends and short on assets with downward trends[26][35][36] - **Commodity Cross-Sectional Inventory Model**: This model is based on the inventory factor to reflect changes in the domestic commodity fundamentals. It dynamically goes long on assets with declining inventory and short on assets with increasing inventory[26][40][43] - **Evaluation of Models**: The Commodity Term Structure Model is noted for its strong performance, achieving a new high in net value during backtesting. The Time-Series Momentum Model has shown weaker performance recently, while the Cross-Sectional Inventory Model has demonstrated moderate gains[25][35][40] - **Backtesting Results**: - Commodity Term Structure Model: Two-week return of 2.31%, year-to-date return of 7.46%[30][33][34] - Commodity Time-Series Momentum Model: Two-week return of -0.38%, year-to-date return of -3.19%[35][36][39] - Commodity Cross-Sectional Inventory Model: Two-week return of 0.98%, year-to-date return of 5.43%[40][43][44]