商品量化CTA周度跟踪

Quantitative Models and Construction - Model Name: Composite Signal Model Construction Idea: The model integrates multiple factors, including supply, demand, inventory, and price spread, to generate a composite signal for market positioning. [5][9][11] Construction Process: - Supply Factor: Tracks production metrics such as glass production and iron ore shipment volumes. Signals are derived based on production trends and their deviations from historical averages. [9][10][11] - Demand Factor: Incorporates metrics like commodity house sales and battery prices to assess downstream consumption trends. [9][11] - Inventory Factor: Monitors stock levels at ports and warehouses, such as iron ore inventory and glass inventory changes. [10][11] - Price Spread Factor: Evaluates price differences across regions or timeframes, e.g., methanol price spreads and freight costs. [8][10][11] Evaluation: The model effectively captures multi-dimensional market dynamics but may face challenges in volatile environments. [5][9][11] Quantitative Factors and Construction - Factor Name: Supply Factor Construction Idea: Measures production activity and its impact on market supply. [5][9][10] Construction Process: - Glass production trends and iron ore shipment volumes are analyzed. [9][10] - Formula: $ \text{Supply Signal} = \text{Production Change} - \text{Historical Average} $ Evaluation: Provides reliable insights into supply-side pressures but may lag during sudden disruptions. [9][10] - Factor Name: Demand Factor Construction Idea: Tracks downstream consumption metrics to gauge market demand. [5][9][11] Construction Process: - Metrics include commodity house sales and battery prices. [9][11] - Formula: $ \text{Demand Signal} = \text{Consumption Change} - \text{Historical Average} $ Evaluation: Captures demand trends effectively but may underperform in rapidly shifting consumption patterns. [9][11] - Factor Name: Inventory Factor Construction Idea: Monitors stock levels to assess market balance. [5][9][10] Construction Process: - Tracks port and warehouse inventory changes, e.g., iron ore and glass inventory. [10][11] - Formula: $ \text{Inventory Signal} = \text{Current Inventory} - \text{Historical Average} $ Evaluation: Useful for identifying market imbalances but sensitive to reporting delays. [10][11] - Factor Name: Price Spread Factor Construction Idea: Evaluates price differences across regions or timeframes to identify arbitrage opportunities. [5][8][10] Construction Process: - Analyzes methanol price spreads and freight costs. [8][10] - Formula: $ \text{Price Spread Signal} = \text{Regional Price Difference} - \text{Historical Spread} $ Evaluation: Effective for arbitrage identification but may struggle in highly volatile markets. [8][10] Backtesting Results - Composite Signal Model: - Weekly Return: 0.18% [5] - Monthly Return: -0.08% [5] - Supply Factor: - Weekly Return: 1.20% [9] - Monthly Return: 2.66% [9] - Demand Factor: - Weekly Return: 0.42% [5] - Monthly Return: -0.29% [9] - Inventory Factor: - Weekly Return: 0.00% [9] - Monthly Return: 0.91% [9] - Price Spread Factor: - Weekly Return: 1.10% [9] - Monthly Return: 2.22% [9] - Profit Factor: - Weekly Return: 0.42% [9] - Monthly Return: 3.23% [9]