Quantitative Models and Construction Methods Model 1: Fund Selection Strategy Based on Trading Motivation Factor and Stock Price Difference Income Factor - **Model Name**: Fund Selection Strategy Based on Trading Motivation Factor and Stock Price Difference Income Factor - **Construction Idea**: The strategy aims to select funds with high stock price difference income, active trading motivation, and low possibility of performance dressing[2] - **Construction Process**: - The strategy combines the trading motivation factor and the stock price difference income factor - The trading motivation factor is constructed by classifying the trading motivations of funds[23] - The stock price difference income factor is derived from the stock price difference income in the fund's income statement[23] - The strategy adopts a semi-annual rebalancing approach, rebalancing at the end of March and August each year[23] - **Evaluation**: The strategy significantly outperformed the Wind Partial Equity Hybrid Fund Index in January 2026[2] Model 2: Fund Selection Strategy Based on Fund Manager's Trading Uniqueness - **Model Name**: Fund Selection Strategy Based on Fund Manager's Trading Uniqueness - **Construction Idea**: The strategy aims to capture the unique trading patterns of fund managers to generate excess returns[3] - **Construction Process**: - Construct a network based on the detailed holdings and transactions of fund managers[31] - Develop an indicator to measure the uniqueness of fund managers' trading[31] - The strategy adopts a semi-annual rebalancing approach, rebalancing at the beginning of April and September each year[31] - **Evaluation**: The strategy outperformed the Wind Partial Equity Hybrid Fund Index in January 2026[3] Model 3: Industry Theme ETF Rotation Strategy Based on Application Information - **Model Name**: Industry Theme ETF Rotation Strategy Based on Application Information - **Construction Idea**: The strategy aims to select industry theme ETFs similar to the applied ETFs to capture market investment hotspots[4] - **Construction Process**: - Conduct event-driven research on the entire issuance process of funds[36] - Construct the industry theme application similarity factor (T+1) based on the information disclosed during the application material public stage[36] - The strategy adopts a monthly rebalancing approach, with a transaction fee rate of 0.1% per side[36] - **Evaluation**: The strategy significantly outperformed the CSI 800 Index in January 2026[4] Model Backtesting Results Fund Selection Strategy Based on Trading Motivation Factor and Stock Price Difference Income Factor - **Monthly Return**: 10.96%[27] - **Annualized Return**: 11.56%[27] - **Annualized Volatility**: 21.60%[27] - **Sharpe Ratio**: 0.54[27] - **Maximum Drawdown**: 48.39%[27] - **Annualized Excess Return**: 3.87%[27] - **Excess Maximum Drawdown**: 19.22%[27] - **Information Ratio (IR)**: 0.64[27] - **Monthly Excess Return**: 3.60%[27] Fund Selection Strategy Based on Fund Manager's Trading Uniqueness - **Monthly Return**: 8.03%[35] - **Annualized Return**: 14.26%[35] - **Annualized Volatility**: 19.47%[35] - **Sharpe Ratio**: 0.73[35] - **Maximum Drawdown**: 37.26%[35] - **Annualized Excess Return**: 5.70%[35] - **Excess Maximum Drawdown**: 10.84%[35] - **Information Ratio (IR)**: 1.10[35] - **Monthly Excess Return**: 0.86%[35] Industry Theme ETF Rotation Strategy Based on Application Information - **Monthly Return**: 22.66%[40] - **Annualized Return**: 22.45%[40] - **Annualized Volatility**: 21.39%[40] - **Sharpe Ratio**: 1.05[40] - **Maximum Drawdown**: 34.89%[43] - **Annualized Excess Return**: 13.84%[43] - **Excess Maximum Drawdown**: 19.07%[43] - **Information Ratio (IR)**: 0.76[43] - **Monthly Excess Return**: 18.18%[43]