Quantitative Models and Construction Methods 1. Model Name: Micro-Cap Timing Model - **Model Construction Idea**: This model uses mid-term risk warning indicators to monitor market timing for micro-cap stocks, focusing on both market sentiment and fundamental indicators[29] - **Model Construction Process**: 1. Two key indicators are used: - **10-Year Treasury Yield YoY**: Measures the year-over-year change in 10-year treasury yields - **Volatility Crowding YoY**: Reflects market trading sentiment through changes in volatility 2. Risk thresholds are set for these indicators: - 10-Year Treasury Yield YoY threshold: 0.3 - Volatility Crowding YoY threshold: 0 3. If either indicator exceeds its threshold, a risk warning signal is triggered. Conversely, if both indicators fall below their thresholds, the risk warning is lifted[29][30][36] - **Model Evaluation**: The model currently does not trigger any risk warnings, suggesting a stable environment for holding micro-cap stocks[29] --- Model Backtesting Results 1. Micro-Cap Timing Model - **10-Year Treasury Yield YoY**: -28.69% (below the risk threshold of 0.3)[30] - **Volatility Crowding YoY**: -50.09% (below the risk threshold of 0)[30] --- Quantitative Factors and Construction Methods 1. Factor Name: Value Factor - **Factor Construction Idea**: Captures undervaluation by comparing financial metrics to market prices[51] - **Factor Construction Process**: 1. **Key Metrics**: - **SP_TTM**: Revenue over the past 12 months divided by the latest market capitalization - **EP_FY0**: Current annual report's expected net profit divided by the latest market capitalization - **Sales2EV**: Revenue over the past 12 months divided by enterprise value 2. These metrics are aggregated to form the value factor, with higher values indicating greater undervaluation[51] 2. Factor Name: Size Factor - **Factor Construction Idea**: Measures the impact of company size on stock performance[51] - **Factor Construction Process**: 1. **Key Metrics**: - **LN_MktCap**: Natural logarithm of free-float market capitalization - **BP_LR**: Book value per share divided by the latest market price - **EP_FTTM**: Forward 12-month expected earnings divided by the latest market price 2. These metrics are combined to represent the size factor, with smaller companies typically expected to outperform[51] 3. Factor Name: Volatility Factor - **Factor Construction Idea**: Captures the impact of price and return volatility on stock performance[51] - **Factor Construction Process**: 1. **Key Metrics**: - **IV_CAPM**: Residual volatility from the CAPM model - **IV_FF**: Residual volatility from the Fama-French three-factor model - **Price_Chg20D**: 20-day price change 2. These metrics are aggregated to form the volatility factor, with lower volatility generally preferred[51] 4. Factor Name: Reversal Factor - **Factor Construction Idea**: Exploits the tendency of stocks to revert to their mean after extreme movements[51] - **Factor Construction Process**: 1. **Key Metrics**: - **Price_Chg40D**: 40-day price change - **Price_Chg60D**: 60-day price change - **Price_Chg120D**: 120-day price change 2. These metrics are aggregated to form the reversal factor, with negative values indicating stronger reversal potential[51] 5. Factor Name: Convertible Bond Valuation Factor - **Factor Construction Idea**: Evaluates convertible bonds based on their valuation and relationship with underlying stocks[44] - **Factor Construction Process**: 1. **Key Metrics**: - **Parity Premium**: Difference between the convertible bond price and its parity value - **Underlying Stock Factors**: Includes growth, quality, and valuation metrics of the underlying stock 2. These metrics are combined to assess the attractiveness of convertible bonds[44][49] --- Factor Backtesting Results 1. Value Factor - **IC Mean**: Positive across multiple stock pools, with notable performance in the financial and real estate sectors[39][40] - **Multi-Long-Short Returns**: Positive returns observed in the past week[39][42] 2. Size Factor - **IC Mean**: Strong performance in the CSI 300 stock pool[39][41] - **Multi-Long-Short Returns**: Expected to improve as market risk appetite increases[40] 3. Volatility Factor - **IC Mean**: Stable positive returns in the A-share stock pool[39][42] - **Multi-Long-Short Returns**: Demonstrated consistent performance in recent weeks[39][42] 4. Reversal Factor - **IC Mean**: Outperformed in the CSI 500 and CSI 1000 stock pools[39][41] - **Multi-Long-Short Returns**: Positive returns observed in the past week[39][42] 5. Convertible Bond Valuation Factor - **Multi-Long-Short Returns**: Positive returns for value and valuation factors in the convertible bond market[44][48]

Investment Rating - The report maintains a positive outlook on the micro-cap stock index, indicating a favorable investment environment following regulatory changes [3][13]. Core Insights - The recent modification of the "Major Asset Restructuring Management Measures" by the China Securities Regulatory Commission allows for cross-border mergers and acquisitions, which is seen as a significant positive for micro-cap stocks [3][13]. - The micro-cap stock index has shown strong performance, ranking first among 38 broad indices over the past month with a 12.51% increase and a 68.6% increase over the past year [4][14][24]. Summary by Sections 1. Overview of Micro-Cap Stock Index Performance - The micro-cap stock index increased by 1.58% over the past week, ranking 5th among 38 indices [4][14]. - Over the past month, the index rose by 12.51%, ranking 1st [4][19]. - In the last quarter, the index saw a 17.17% increase, also ranking 1st [4][20]. - The annual performance shows a significant increase of 68.6%, ranking 2nd [4][22][24]. 2. Factor Performance of Micro-Cap Stocks - The top-performing factors this week include standardized expected earnings (0.154), leverage (0.127), and quarterly net profit growth (0.102) [5][14][28]. - The bottom-performing factors include momentum (-0.08) and growth factors (-0.032) [5][14][28]. 3. Diffusion Index Observations - The diffusion index remains high but has shown a slight decline, indicating potential sell signals if the trend continues [6][15][33]. - The first threshold method triggered a sell signal on May 8, 2025, while the delayed threshold method provided a buy signal on April 22, 2025 [6][15][36][40]. 4. Calendar Effects - The micro-cap stock index has shown positive average returns on Tuesdays and Thursdays, while Mondays and Wednesdays have negative average returns [7][16]. - The index achieved an 84% win rate in February and a 90% win rate in March and May [7][16]. 5. Micro-Cap Stock Related Fund Performance - The highest-performing micro-cap fund this week yielded 2.79%, while the lowest yielded 0.35% [16][43].

- The diffusion index model triggered a right-side buy signal at the close of trading on Tuesday, April 25, 2025 [6][14][33] - The first threshold method triggered a full position signal at the close of trading on April 7, 2025 [36] - The delayed threshold method gave an open position signal at the close of trading on April 22, 2025 [40] - The dual moving average method gave a short position signal at the close of trading on March 12, 2025 [41] Model Testing Results - Diffusion index model, weekly return 2.52% [43] - First threshold method, signal value 0.0225 [36] - Delayed threshold method, signal value not specified [40] - Dual moving average method, signal value not specified [41] Quantitative Factors and Construction - Factor name: Momentum factor; Construction idea: Based on the stock's price momentum; Construction process: Calculated using the rank IC method, weekly rank IC value 0.221, historical average -0.003 [5][13][27] - Factor name: Logarithmic market capitalization factor; Construction idea: Based on the logarithm of the stock's market capitalization; Construction process: Calculated using the rank IC method, weekly rank IC value 0.196, historical average -0.032 [5][13][27] - Factor name: Non-linear market capitalization factor; Construction idea: Based on the non-linear transformation of the stock's market capitalization; Construction process: Calculated using the rank IC method, weekly rank IC value 0.196, historical average -0.032 [5][13][27] - Factor name: Single-quarter ROE factor; Construction idea: Based on the stock's return on equity for a single quarter; Construction process: Calculated using the rank IC method, weekly rank IC value 0.19, historical average 0.025 [5][13][27] - Factor name: PE TTM reciprocal factor; Construction idea: Based on the reciprocal of the stock's price-to-earnings ratio trailing twelve months; Construction process: Calculated using the rank IC method, weekly rank IC value 0.189, historical average 0.02 [5][13][27] Factor Testing Results - Momentum factor, weekly rank IC value 0.221, historical average -0.003 [5][13][27] - Logarithmic market capitalization factor, weekly rank IC value 0.196, historical average -0.032 [5][13][27] - Non-linear market capitalization factor, weekly rank IC value 0.196, historical average -0.032 [5][13][27] - Single-quarter ROE factor, weekly rank IC value 0.19, historical average 0.025 [5][13][27] - PE TTM reciprocal factor, weekly rank IC value 0.189, historical average 0.02 [5][13][27]