2025 年 10 月 31 日 金融工程研究团队 魏建榕（首席分析师） 证书编号：S0790519120001 张 翔（分析师） 证书编号：S0790520110001 傅开波（分析师） 证书编号：S0790520090003 高 鹏（分析师） 证书编号：S0790520090002 苏俊豪（分析师） 证书编号：S0790522020001 胡亮勇（分析师） 证书编号：S0790522030001 王志豪（分析师） 证书编号：S0790522070003 盛少成（分析师） 证书编号：S0790523060003 苏 良（分析师） 证书编号：S0790523060004 何申昊（研究员） 证书编号：S0790122080094 蒋 韬（研究员） 证书编号：S0790123070037 相关研究报告 《10 月转债配置：转债估值偏贵，看好 偏 股 低 估 风 格 — 金 融 工 程 定 期 》 -2025.10.17 《量化产品季度点评：300&500 增强 Q3 超额回撤，公募红利量化表现优异—开 源量化评论（113）》-2025.10.11 《有色金属板块的资金行为监测—金融 工程定期》-2025.10.1 ...

金融工程研究团队 2025 年 09 月 26 日 魏建榕（首席分析师） 证书编号：S0790519120001 张 翔（分析师） 证书编号：S0790520110001 傅开波（分析师） 证书编号：S0790520090003 高 鹏（分析师） 证书编号：S0790520090002 苏俊豪（分析师） 证书编号：S0790522020001 胡亮勇（分析师） 证书编号：S0790522030001 王志豪（分析师） 盛少成（分析师） 证书编号：S0790523060003 苏 良（分析师） 证书编号：S0790523060004 何申昊（研究员） 证书编号：S0790122080094 蒋 韬（研究员） 证书编号：S0790123070037 相关研究报告 《商品择时及其在资产配置中的应用— 大类资产配置研究系列（13）》-2025.9.19 《9 月转债配置：转债估值偏贵，看好 偏 股 低 估 风 格 — 金 融 工 程 定 期 》 -2025.9.16 《基于港交所 CCASS 数据的港股投资 策略—开源量化评论（112）》-2025.9.14 开源交易行为因子绩效月报（2025 年 9 月） ——金融 ...

Quantitative Models and Construction Methods 1. Model Name: Convertible Bond Random Forest Strategy - **Model Construction Idea**: Utilizes the Random Forest machine learning method to identify convertible bonds with potential for excess returns by leveraging decision trees[16][17] - **Model Construction Process**: 1. Data preprocessing and feature engineering to prepare convertible bond datasets 2. Training a Random Forest model with historical data to identify patterns of excess return potential 3. Selecting bonds with the highest predicted scores for portfolio construction 4. Weekly rebalancing of the portfolio based on updated predictions[17] - **Model Evaluation**: Demonstrated strong performance in generating excess returns, indicating high predictive accuracy[16] 2. Model Name: Multi-Dimensional Timing Model - **Model Construction Idea**: Combines macro, meso, micro, and derivative signals to create a four-dimensional non-linear timing model for market positioning[18][19] - **Model Construction Process**: 1. Macro signals: Derived from liquidity, interest rates, credit, economic growth, and exchange rates 2. Meso signals: Based on industry-level business cycle indicators 3. Micro signals: Captures structural risks using valuation, risk premium, volatility, and liquidity factors 4. Derivative signals: Generated from the basis of stock index futures 5. Aggregation: Signals are synthesized into a composite timing signal[18][19][24] - **Model Evaluation**: Effective in identifying market trends and providing actionable signals, with the latest signal indicating a bullish stance[19][24] 3. Model Name: Industry Rotation Strategy 2.0 - **Model Construction Idea**: Constructs an industry rotation strategy based on economic quadrants and multi-dimensional industry style factors[69] - **Model Construction Process**: 1. Define economic quadrants using corporate earnings and credit conditions 2. Develop industry style factors such as expected business climate, earnings surprises, momentum, valuation bubbles, and inflation beta 3. Test factor effectiveness within each quadrant 4. Allocate to high-expected-return industries based on factor signals[69][71] - **Model Evaluation**: Demonstrates strong adaptability to the A-share market, with annualized excess returns of 9.44% (non-exclusion version) and 10.14% (double-exclusion version)[71] 4. Model Name: Genetic Programming Index Enhancement Models - **Model Construction Idea**: Uses genetic programming to discover and optimize stock selection factors for index enhancement strategies[88][93][97] - **Model Construction Process**: 1. Stock pools: Defined for CSI 300, CSI 500, CSI 1000, and CSI All Share indices 2. Training: Genetic programming generates initial factor populations and iteratively evolves them through multiple generations 3. Factor selection: Top-performing factors are combined into a composite score 4. Portfolio construction: Selects top 10% of stocks within each industry based on scores, with weekly rebalancing[88][93][97][102] - **Model Evaluation**: - CSI 300: Annualized excess return of 17.91%, Sharpe ratio of 1.05[91] - CSI 500: Annualized excess return of 11.78%, Sharpe ratio of 0.85[95] - CSI 1000: Annualized excess return of 17.97%, Sharpe ratio of 0.93[98] - CSI All Share: Annualized excess return of 24.84%, Sharpe ratio of 1.33[103] --- Model Backtest Results 1. Convertible Bond Random Forest Strategy - Weekly excess return: 0.64%[16] 2. Multi-Dimensional Timing Model - Latest composite signal: Bullish (1)[19][24] 3. Industry Rotation Strategy 2.0 - Annualized excess return (non-exclusion version): 9.44% - Annualized excess return (double-exclusion version): 10.14%[71] 4. Genetic Programming Index Enhancement Models - CSI 300: - Annualized excess return: 17.91% - Sharpe ratio: 1.05[91] - CSI 500: - Annualized excess return: 11.78% - Sharpe ratio: 0.85[95] - CSI 1000: - Annualized excess return: 17.97% - Sharpe ratio: 0.93[98] - CSI All Share: - Annualized excess return: 24.84% - Sharpe ratio: 1.33[103] --- Quantitative Factors and Construction Methods 1. Factor Name: Industry Business Climate Index 2.0 - **Factor Construction Idea**: Tracks industry fundamentals by analyzing revenue, pricing, and cost dynamics[27] - **Factor Construction Process**: 1. Analyze industry revenue and cost structures 2. Calculate daily market-cap-weighted industry indices 3. Aggregate indices into a composite business climate index[27][30] - **Factor Evaluation**: Demonstrates predictive power for A-share earnings expansion cycles[28] 2. Factor Name: Barra CNE6 Style Factors - **Factor Construction Idea**: Evaluates market performance using 9 primary and 20 secondary style factors, including size, volatility, momentum, quality, value, and growth[45] - **Factor Construction Process**: 1. Calculate factor returns for each style factor 2. Aggregate factor performance to assess market trends[45][46] - **Factor Evaluation**: Size factor performed well during the week, while volatility factor underperformed[46] 3. Factor Name: Industry Rotation Factors - **Factor Construction Idea**: Captures industry rotation dynamics using factors like expected business climate, earnings surprises, momentum, and valuation bubbles[69] - **Factor Construction Process**: 1. Define and calculate individual factors 2. Test factor effectiveness within economic quadrants 3. Combine factors for industry allocation[69] - **Factor Evaluation**: Demonstrates strong historical performance, with factors like expected business climate and momentum showing significant returns[57][59] --- Factor Backtest Results 1. Industry Business Climate Index 2.0 - Current value: 0.913 - Excluding financials: 1.288[28] 2. Barra CNE6 Style Factors - Size factor: Strong performance during the week[46] 3. Industry Rotation Factors - Historical annualized returns: - Expected business climate: 0.40% - Momentum: -0.95% - Valuation beta: 2.37%[57]

2025 年 08 月 29 日 金融工程研究团队 魏建榕（首席分析师） 证书编号：S0790519120001 张 翔（分析师） 证书编号：S0790520110001 傅开波（分析师） 证书编号：S0790520090003 高 鹏（分析师） 证书编号：S0790520090002 苏俊豪（分析师） 证书编号：S0790522020001 胡亮勇（分析师） 证书编号：S0790522030001 王志豪（分析师） 盛少成（分析师） 证书编号：S0790523060003 苏 良（分析师） 证书编号：S0790523060004 何申昊（研究员） 证书编号：S0790122080094 蒋 韬（研究员） 证书编号：S0790123070037 相关研究报告 开源交易行为因子绩效月报（2025 年 8 月） ——金融工程定期 | 魏建榕（分析师） | 高鹏（分析师） | 盛少成（分析师） | | --- | --- | --- | | weijianrong@kysec.cn | gaopeng@kysec.cn | shengshaocheng@kysec.cn | | 证书编号：S079051912000 ...

- Model Name: Barra Style Factors; Model Construction Idea: Measure the performance of common Barra style factors in April 2025; Model Construction Process: Calculate the returns of various factors such as market capitalization, book-to-market ratio, growth, and earnings expectations; Model Evaluation: Provides insights into the performance of different style factors in the market[4][14] - Factor Name: Ideal Reversal Factor; Factor Construction Idea: Identify the strongest reversal days based on the average transaction amount per trade; Factor Construction Process: 1. Retrieve the past 20 days of data for the selected stock 2. Calculate the average transaction amount per trade for each day 3. Sum the returns of the top 10 days with the highest average transaction amount, denoted as M_high 4. Sum the returns of the bottom 10 days with the lowest average transaction amount, denoted as M_low 5. Calculate the Ideal Reversal Factor as M = M_high - M_low[5][46][49] - Factor Name: Smart Money Factor; Factor Construction Idea: Identify the participation of smart money in trading based on minute-level price and volume data; Factor Construction Process: 1. Retrieve the past 10 days of minute-level data for the selected stock 2. Construct the indicator $ S_t = \frac{|R_t|}{V_t^{0.25}} $, where $ R_t $ is the return at minute t and $ V_t $ is the volume at minute t 3. Sort the minute data by $ S_t $ in descending order and select the top 20% of minutes by cumulative volume as smart money trades 4. Calculate the volume-weighted average price (VWAP) of smart money trades, denoted as VWAP_smart 5. Calculate the VWAP of all trades, denoted as VWAP_all 6. Calculate the Smart Money Factor as $ Q = \frac{VWAP_{smart}}{VWAP_{all}} $[5][47] - Factor Name: APM Factor; Factor Construction Idea: Measure the difference in stock price behavior between morning (or overnight) and afternoon sessions; Factor Construction Process: 1. Retrieve the past 20 days of data for the selected stock 2. Record the overnight and afternoon returns for both the stock and the index 3. Perform a regression of the form $ r_t = \alpha + \beta R_t + \epsilon_t $ to obtain residuals 4. Calculate the difference between overnight and afternoon residuals 5. Construct the statistic $ \text{stat} = \frac{\mu(\delta_t)}{\sigma(\delta_t)/\sqrt{N}} $ 6. Regress the statistic against the momentum factor to obtain the APM Factor[5][48][50] - Factor Name: Ideal Amplitude Factor; Factor Construction Idea: Measure the difference in amplitude information between high and low price states; Factor Construction Process: 1. Retrieve the past 20 days of data for the selected stock 2. Calculate the daily amplitude as (highest price/lowest price - 1) 3. Calculate the average amplitude for the top 25% of days with the highest closing prices, denoted as V_high 4. Calculate the average amplitude for the bottom 25% of days with the lowest closing prices, denoted as V_low 5. Calculate the Ideal Amplitude Factor as V = V_high - V_low[5][51] Model and Factor Performance - Barra Style Factors: Market Capitalization Factor return: 0.09%, Book-to-Market Ratio Factor return: 0.11%, Growth Factor return: -0.19%, Earnings Expectations Factor return: -0.02%[4][14] - Ideal Reversal Factor: IC: -0.051, rankIC: -0.061, IR: 2.55, Long-Short Monthly Win Rate: 78.5%, April 2025 Long-Short Return: 0.89%, Last 12 Months Long-Short Monthly Win Rate: 66.7%[6][16] - Smart Money Factor: IC: -0.038, rankIC: -0.061, IR: 2.78, Long-Short Monthly Win Rate: 82.5%, April 2025 Long-Short Return: 0.89%, Last 12 Months Long-Short Monthly Win Rate: 100.0%[6][21] - APM Factor: IC: 0.030, rankIC: 0.034, IR: 2.32, Long-Short Monthly Win Rate: 77.6%, April 2025 Long-Short Return: -0.27%, Last 12 Months Long-Short Monthly Win Rate: 75.0%[6][25] - Ideal Amplitude Factor: IC: -0.054, rankIC: -0.073, IR: 3.04, Long-Short Monthly Win Rate: 83.9%, April 2025 Long-Short Return: 2.52%, Last 12 Months Long-Short Monthly Win Rate: 83.3%[6][30] - Composite Trading Behavior Factor: IC: 0.068, rankIC: 0.092, IR: 3.36, Long-Short Monthly Win Rate: 82.2%, April 2025 Long-Short Return: 0.99%, Last 12 Months Long-Short Monthly Win Rate: 83.3%[6][35]