Quantitative Models and Construction Methods 1. Model Name: Short-term Quantitative Timing Model - **Model Construction Idea**: This model integrates signals from valuation, liquidity, fundamentals, and sentiment to generate short-term market timing signals[17][18] - **Model Construction Process**: - **Valuation Signal**: Based on PE and PB percentile rankings over the past 5 years. A higher percentile indicates optimism, while a lower percentile indicates caution[17][21] - **Liquidity Signal**: Derived from indicators such as money market rates, exchange rate expectations, and net financing. Favorable liquidity conditions generate optimistic signals[18][21] - **Fundamental Signal**: Includes metrics like manufacturing PMI, long-term loan growth, and M1 growth. Stronger fundamentals lead to optimistic signals[17][21] - **Sentiment Signal**: Combines metrics like trading volume, turnover rate, and market volatility. Higher sentiment scores indicate optimism[18][21] - **Signal Aggregation**: Signals from the four dimensions are aggregated to determine the overall market timing signal (optimistic, neutral, or cautious)[17][18] - **Model Evaluation**: The model demonstrates strong predictive power for short-term market movements, with significant outperformance over the benchmark strategy[19][22] 2. Model Name: Growth-Value Style Rotation Model - **Model Construction Idea**: This model identifies the relative attractiveness of growth versus value styles based on fundamentals, valuation, and sentiment[27][28] - **Model Construction Process**: - **Fundamental Signal**: Factors include profit cycle slope, interest rate cycle level, and credit cycle changes. Favorable conditions for growth or value are identified based on these metrics[27][29] - **Valuation Signal**: PE and PB valuation spreads between growth and value are analyzed. A narrowing spread favors growth, while a widening spread favors value[27][29] - **Sentiment Signal**: Metrics like turnover and volatility differences between growth and value are used. Lower turnover favors value, while higher volatility favors balance[28][29] - **Signal Aggregation**: Signals from the three dimensions are combined to determine the allocation between growth and value styles[27][28] - **Model Evaluation**: The strategy has historically outperformed a balanced allocation, though it underperformed in certain years like 2014 and 2020[28][31] 3. Model Name: Small-Cap vs. Large-Cap Style Rotation Model - **Model Construction Idea**: This model evaluates the relative attractiveness of small-cap versus large-cap styles based on fundamentals, valuation, and sentiment[32][33] - **Model Construction Process**: - **Fundamental Signal**: Similar to the growth-value model, factors include profit cycle slope, interest rate cycle level, and credit cycle changes. Favorable conditions for small-cap or large-cap are identified[32][34] - **Valuation Signal**: PE and PB valuation spreads between small-cap and large-cap are analyzed. A narrowing spread favors large-cap, while a widening spread favors small-cap[33][34] - **Sentiment Signal**: Metrics like turnover and volatility differences between small-cap and large-cap are used. Higher turnover favors small-cap, while lower volatility favors large-cap[33][34] - **Signal Aggregation**: Signals from the three dimensions are combined to determine the allocation between small-cap and large-cap styles[32][33] - **Model Evaluation**: The strategy has historically delivered significant outperformance over a balanced allocation, though it underperformed in certain years like 2020[33][35] 4. Model Name: Four-Style Rotation Model - **Model Construction Idea**: This model combines the conclusions of the growth-value and small-cap-large-cap models to allocate across four styles: small-cap growth, small-cap value, large-cap growth, and large-cap value[38][39] - **Model Construction Process**: - **Signal Aggregation**: The allocation is determined by combining the signals from the growth-value and small-cap-large-cap models. For example, the current allocation is 37.5% small-cap growth, 12.5% small-cap value, 37.5% large-cap growth, and 12.5% large-cap value[38][39] - **Model Evaluation**: The strategy has consistently outperformed a balanced allocation, with significant annualized excess returns. However, it underperformed in years like 2014 and 2020[38][39] --- Model Backtesting Results 1. Short-term Quantitative Timing Model - **Annualized Return**: 16.16% - **Annualized Volatility**: 14.80% - **Maximum Drawdown**: 27.70% - **Sharpe Ratio**: 0.9509 - **Monthly Win Rate**: 67.79%[19][22][24] 2. Growth-Value Style Rotation Model - **Annualized Return**: 11.06% - **Annualized Volatility**: 21.01% - **Maximum Drawdown**: 43.07% - **Sharpe Ratio**: 0.5101 - **Monthly Win Rate**: 57.43%[28][31] 3. Small-Cap vs. Large-Cap Style Rotation Model - **Annualized Return**: 11.84% - **Annualized Volatility**: 22.91% - **Maximum Drawdown**: 50.65% - **Sharpe Ratio**: 0.5168 - **Monthly Win Rate**: 60.14%[33][35] 4. Four-Style Rotation Model - **Annualized Return**: 12.65% - **Annualized Volatility**: 21.75% - **Maximum Drawdown**: 47.91% - **Sharpe Ratio**: 0.5656 - **Monthly Win Rate**: 58.78%[38][39]