多因子选股周报 成长因子表现出色，四大指增组合年内超额均超 10% 证券研究报告 | 2025年08月23日 核心观点 金融工程周报 国信金工指数增强组合表现跟踪 因子表现监控 以沪深 300 指数为选股空间。最近一周，标准化预期外收入、一年动量、单 季营收同比增速等因子表现较好，而 EPTTM、单季 EP、一个月换手等因子 表现较差。 以中证 500 指数为选股空间。最近一周，EPTTM 一年分位点、高管薪酬、 DELTAROA 等因子表现较好，而预期 EPTTM、EPTTM、单季 EP 等因子 表现较差。 以中证 1000 指数为选股空间。最近一周，标准化预期外收入、三个月反转、 单季营收同比增速等因子表现较好，而一个月反转、预期 EPTTM、股息率 等因子表现较差。 以中证 A500 指数为选股空间。最近一周，单季营收同比增速、三个月反转、 一年动量等因子表现较好，而单季 EP、EPTTM、预期 EPTTM 等因子表现 较差。 以公募重仓指数为选股空间。最近一周，一年动量、单季营利同比增速、单 季营收同比增速等因子表现较好，而单季 EP、预期 EPTTM、EPTTM 等因 子表现较差。 公募基金指数增强 ...

- The report tracks the performance of Guosen JinGong's index enhancement portfolios and public fund index enhancement products, as well as monitors the performance of common stock selection factors across different sample spaces[11][12][15] - Guosen JinGong's index enhancement portfolios are constructed based on three main components: return prediction, risk control, and portfolio optimization. These portfolios are benchmarked against indices such as CSI 300, CSI 500, CSI 1000, and CSI A500[12][14] - The report introduces the concept of Maximized Factor Exposure (MFE) portfolios to test the effectiveness of single factors under real-world constraints. The optimization model maximizes single-factor exposure while controlling for style, industry, stock weight deviations, and other constraints[41][42][43] - The optimization model for MFE portfolios is expressed as: $\begin{array}{ll}max&f^{T}\ w\\ s.t.&s_{l}\leq X(w-w_{b})\leq s_{h}\\ &h_{l}\leq H(w-w_{b})\leq h_{h}\\ &w_{l}\leq w-w_{b}\leq w_{h}\\ &b_{l}\leq B_{b}w\leq b_{h}\\ &\mathbf{0}\leq w\leq l\\ &\mathbf{1}^{T}\ w=1\end{array}$ where `f` represents factor values, `w` is the stock weight vector, and constraints include style exposure, industry exposure, stock weight deviations, and component stock weight limits[41][42] - The report tracks the performance of single-factor MFE portfolios across different sample spaces, including CSI 300, CSI 500, CSI 1000, CSI A500, and public fund heavy positions index. Factors are evaluated based on their excess returns relative to benchmarks[15][18][26] - Common stock selection factors are categorized into valuation, reversal, growth, profitability, liquidity, company governance, and analyst dimensions. Examples include BP (Book-to-Price), ROA (Return on Assets), and one-year momentum[16][17] - In the CSI 300 sample space, factors such as single-season ROA, standardized unexpected income, and standardized unexpected earnings performed well recently, while factors like one-month volatility and three-month volatility performed poorly[19] - In the CSI 500 sample space, factors such as one-year momentum and standardized unexpected earnings showed strong performance recently, while factors like one-month turnover and three-month volatility underperformed[21] - In the CSI 1000 sample space, factors such as one-year momentum and standardized unexpected earnings performed well recently, while factors like BP and single-season SP (Sales-to-Price) performed poorly[23] - In the CSI A500 sample space, factors such as DELTAROA (Change in ROA) and standardized unexpected income performed well recently, while factors like three-month volatility and one-month turnover performed poorly[25] - In the public fund heavy positions index sample space, factors such as one-year momentum and DELTAROA performed well recently, while factors like one-month turnover and three-month turnover underperformed[27] - Public fund index enhancement products are tracked for their excess returns relative to benchmarks. For CSI 300 products, recent weekly excess returns ranged from -1.41% to 1.91%, with a median of -0.09%[32] - For CSI 500 products, recent weekly excess returns ranged from -2.05% to 0.52%, with a median of -0.51%[34] - For CSI 1000 products, recent weekly excess returns ranged from -1.70% to 0.94%, with a median of -0.53%[37] - For CSI A500 products, recent weekly excess returns ranged from -1.10% to 0.71%, with a median of -0.25%[40]

Quantitative Factors and Construction Methods 1. Factor Name: Book-to-Price Ratio (BP) - **Construction Idea**: Measures the valuation of a stock by comparing its book value to its market value [12] - **Construction Process**: - Formula: $ BP = \frac{\text{Current Book Value}}{\text{Current Market Value}} $ [12] - **Evaluation**: BP is a widely used valuation factor, and its positive IC and excess return indicate its effectiveness in identifying undervalued stocks [8][10] 2. Factor Name: BP Three-Year Percentile - **Construction Idea**: Evaluates the relative valuation of a stock over the past three years [12] - **Construction Process**: - Formula: BP Three-Year Percentile = Percentile rank of the current BP within the last three years [12] - **Evaluation**: This factor provides a historical perspective on valuation, which can enhance its predictive power [8][10] 3. Factor Name: Quarterly Gross Profit Margin - **Construction Idea**: Measures profitability by comparing gross profit to sales revenue [12] - **Construction Process**: - Formula: $ \text{Quarterly Gross Profit Margin} = \frac{\text{Quarterly Gross Profit}}{\text{Quarterly Sales Revenue}} $ [12] - **Evaluation**: A positive IC and strong excess return suggest this factor is effective in identifying profitable companies [8][10] 4. Factor Name: 1-Month Average Daily Turnover - **Construction Idea**: Captures liquidity by analyzing the average daily turnover over the past month [12] - **Construction Process**: - Formula: 1-Month Average Daily Turnover = Mean of daily turnover over the last 20 trading days [12] - **Evaluation**: This factor demonstrates strong performance in short-term IC and excess return, indicating its utility in capturing liquidity-driven opportunities [8][10] 5. Factor Name: 1-Month Turnover Volatility - **Construction Idea**: Measures the variability of turnover over the past month to capture liquidity dynamics [12] - **Construction Process**: - Formula: 1-Month Turnover Volatility = Standard deviation of daily turnover over the last 20 trading days [12] - **Evaluation**: High IC and excess return suggest this factor effectively captures liquidity-related anomalies [8][10] 6. Factor Name: Fama-French Three-Factor 1-Month Residual Volatility - **Construction Idea**: Measures idiosyncratic risk by analyzing the residual volatility from the Fama-French three-factor model [12] - **Construction Process**: - Formula: Residual Volatility = Standard deviation of residuals from the regression of daily returns on the Fama-French three factors over the last 20 trading days [12] - **Evaluation**: This factor is effective in capturing risk-related anomalies, as evidenced by its strong IC and excess return [8][10] --- Factor Backtesting Results IC Performance - **BP**: Weekly IC = 5.32%, Monthly IC = 8.04%, Annual IC = 1.81% [8] - **BP Three-Year Percentile**: Weekly IC = 6.26%, Monthly IC = 10.63%, Annual IC = 3.10% [8] - **Quarterly Gross Profit Margin**: Weekly IC = 5.61%, Monthly IC = 3.29%, Annual IC = 0.64% [8] - **1-Month Average Daily Turnover**: Weekly IC = 12.54%, Monthly IC = 14.06%, Annual IC = 2.03% [8] - **1-Month Turnover Volatility**: Weekly IC = 10.49%, Monthly IC = 13.42%, Annual IC = 2.70% [8] - **Fama-French Three-Factor 1-Month Residual Volatility**: Weekly IC = 8.93%, Monthly IC = 13.16%, Annual IC = 3.32% [8] Excess Return Performance - **BP**: Weekly Excess Return = -0.04%, Monthly Excess Return = 1.27%, Annual Excess Return = 2.22% [10] - **BP Three-Year Percentile**: Weekly Excess Return = -0.25%, Monthly Excess Return = 2.07%, Annual Excess Return = 3.63% [10] - **Quarterly Gross Profit Margin**: Weekly Excess Return = 0.68%, Monthly Excess Return = 1.05%, Annual Excess Return = 4.64% [10] - **1-Month Average Daily Turnover**: Weekly Excess Return = 0.64%, Monthly Excess Return = 2.96%, Annual Excess Return = 7.46% [10] - **1-Month Turnover Volatility**: Weekly Excess Return = 0.55%, Monthly Excess Return = 2.95%, Annual Excess Return = 9.61% [10] - **Fama-French Three-Factor 1-Month Residual Volatility**: Weekly Excess Return = 0.29%, Monthly Excess Return = 2.75%, Annual Excess Return = 7.03% [10]