Quantitative Models and Construction Methods 1. Model Name: GRU (Generalized Recurrent Unit) - **Model Construction Idea**: GRU models are used to capture temporal dependencies and patterns in financial data, leveraging recurrent neural network structures to predict stock performance or factor returns[3][4][5] - **Model Construction Process**: The GRU model is trained on historical stock data, incorporating features such as price movements, volume, and other technical indicators. Specific GRU-based models mentioned include: - **open1d**: Focuses on daily opening prices - **close1d**: Focuses on daily closing prices - **barra1d**: Integrates Barra-style risk factors for daily predictions - **barra5d**: Extends Barra-style risk factors to a 5-day horizon[5][6][25] - **Model Evaluation**: GRU models show mixed performance, with some models like open1d performing well, while others like barra1d and barra5d experience significant drawdowns in certain market conditions[5][6][25] --- Model Backtesting Results GRU Model Performance - **open1d**: - Weekly excess return: 1.22% - Monthly excess return: 2.58% - Year-to-date excess return: 6.08%[29][30] - **close1d**: - Weekly excess return: 1.89% - Monthly excess return: 2.91% - Year-to-date excess return: 4.14%[29][30] - **barra1d**: - Weekly excess return: 0.85% - Monthly excess return: 1.50% - Year-to-date excess return: 3.48%[29][30] - **barra5d**: - Weekly excess return: 0.84% - Monthly excess return: 2.25% - Year-to-date excess return: 5.59%[29][30] --- Quantitative Factors and Construction Methods 1. Factor Name: Barra Style Factors - **Factor Construction Idea**: Barra factors are designed to capture systematic risk exposures across various dimensions such as size, value, momentum, and volatility[13][14] - **Factor Construction Process**: - **Beta**: Historical beta of the stock - **Size**: Natural logarithm of total market capitalization - **Momentum**: Weighted average of historical excess returns, combining volatility, cumulative deviation, and residual volatility $ Momentum = 0.74 \cdot \text{Volatility} + 0.16 \cdot \text{Cumulative Deviation} + 0.1 \cdot \text{Residual Volatility} $ - **Volatility**: Weighted average of historical residual return volatilities - **Valuation**: Inverse of price-to-book ratio - **Liquidity**: Weighted average of turnover ratios (monthly, quarterly, yearly) - **Profitability**: Weighted average of analyst forecasted earnings yield, cash flow yield, and other profitability metrics - **Growth**: Weighted average of earnings and revenue growth rates - **Leverage**: Weighted average of market leverage, book leverage, and debt-to-asset ratio[14][15] - **Factor Evaluation**: Barra factors demonstrate varying performance across different market conditions, with some factors like volatility and liquidity showing strong returns, while others like size and growth exhibit weaker performance[15][16] 2. Factor Name: Technical Factors - **Factor Construction Idea**: Technical factors aim to capture price and volume-based patterns, focusing on momentum and volatility metrics[17][20][24] - **Factor Construction Process**: - **Momentum**: Calculated over different time horizons (e.g., 20-day, 60-day, 120-day) - **Volatility**: Measured as the standard deviation of returns over specific periods (e.g., 20-day, 60-day, 120-day) - **Median Deviation**: Captures the median absolute deviation of returns[27] - **Factor Evaluation**: High-momentum and high-volatility stocks generally outperform, but certain periods show negative returns for these factors, especially in the 120-day horizon[17][27] 3. Factor Name: Fundamental Factors - **Factor Construction Idea**: Fundamental factors are derived from financial statements, focusing on profitability, growth, and valuation metrics[17][20][24] - **Factor Construction Process**: - **Static Financial Metrics**: Return on equity (ROE), return on assets (ROA), and profit margins - **Growth Metrics**: Earnings growth, revenue growth, and cash flow growth - **Surprise Metrics**: Earnings and revenue surprises relative to analyst expectations[19][21][23] - **Factor Evaluation**: Growth and surprise factors perform well, while static financial metrics like ROA and ROE show weaker performance in certain periods[19][21][23] --- Factor Backtesting Results Barra Factors - **Volatility**: Weekly return: 0.75%, Monthly return: 2.73% - **Liquidity**: Weekly return: 0.68%, Monthly return: 1.37% - **Size**: Weekly return: -1.45%, Monthly return: -3.60%[15][16] Technical Factors - **20-day Momentum**: Weekly return: -1.81%, Monthly return: -6.16% - **60-day Volatility**: Weekly return: -1.79%, Monthly return: -0.74% - **120-day Momentum**: Weekly return: -1.68%, Monthly return: -0.80%[27] Fundamental Factors - **ROA Growth**: Weekly return: 0.23%, Monthly return: 1.31% - **Earnings Surprise**: Weekly return: 0.20%, Monthly return: 1.11% - **Revenue Growth**: Weekly return: 0.17%, Monthly return: 0.77%[19][21][23]

Investment Rating - The report indicates a cautious view on the market with a focus on small-cap stocks as a preferred investment style [1][12]. Core Insights - The A-share market continued to show volatility with a contraction in trading volume, leading to a cautious sentiment in the market. The liquidity remains loose, favoring small-cap stocks, and a "dividend + small-cap" strategy is suggested for relative returns [1][12]. - Major indices showed mixed performance, with the Shanghai Composite Index rising by 0.76% and the ChiNext Index increasing by 1.38%, while the CSI 500 and CSI 1000 experienced slight declines [1][13]. Summary by Sections Overall Market Performance - The report covers the performance of major indices from May 12 to May 16, 2025, highlighting a mixed outcome with the Shanghai Composite Index up by 0.76% and the ChiNext Index up by 1.38% [1][13]. - The report notes that the liquidity environment remains favorable for small-cap stocks, which are expected to continue outperforming [1][12]. Industry Valuation - As of May 16, 2025, the report categorizes the valuation of major indices as "moderate," while the ChiNext Index is classified as "safe." Specific industries such as non-bank financials, transportation, and utilities are also rated as "safe" [1][18][19]. Fund Flow Tracking - The report indicates that institutional interest was highest in stocks like Anji Technology and Hengda, with significant net outflows from ETFs, particularly in the stock-type ETFs [3][56]. - The report highlights that southbound capital saw a net outflow of 86.85 billion HKD during the tracking period [3][56]. Volatility Analysis - The report discusses the cross-sectional volatility of major indices, noting a decrease in the cross-sectional volatility of the CSI 300, CSI 500, and CSI 1000, indicating a deterioration in the short-term alpha environment [2][37]. - Time series volatility for the CSI 300 increased, suggesting an improvement in the alpha environment, while the CSI 500 and CSI 1000 saw declines [2][41].

点击注册小程序 查看完整报告 特别申明： 本订阅号中所涉及的证券研究信息由光大证券研究所编写，仅面向光大证券专业投资者客户，用作新媒体形势下研究 信息和研究观点的沟通交流。非光大证券专业投资者客户，请勿订阅、接收或使用本订阅号中的任何信息。本订阅号 难以设置访问权限，若给您造成不便，敬请谅解。光大证券研究所不会因关注、收到或阅读本订阅号推送内容而视相 关人员为光大证券的客户。 今 日 聚 焦 【策略】A股牛市历史复盘及前景展望——解密牛市系列之一 基本面回升是牛市形成的核心驱动力，流动性宽松与产业趋势往往形成共振效应。当基本面全面改善时， 通常催生全面牛市，而在基本面结构性改善阶段，若与流动性宽松、产业趋势形成共振，同样可能孕育牛 市。展望未来，基本面修复进程或将呈现温和且渐进的特征，宏微观流动性共振与产业升级有望驱动市场 上涨。 （张宇生/郭磊） 2025-05-18 您可点击今日推送内容的第1条查看 【金工】小市值风格仍占优——金融工程市场跟踪周报20250519 上周A股延续震荡表现，主要宽基指数量能有所收缩。交易情绪方面，伴随量能收缩，截至上周五 （2025.05.16，下同）主要宽基指数量能择时指标 ...

Core Viewpoint - The A-share market continued to show a fluctuating performance with reduced trading volume, indicating a cautious sentiment among investors. The market is experiencing a "profit-taking" state, with net outflows from ETFs [2]. Market Performance Summary - The major indices showed mixed results: the Shanghai Composite Index rose by 0.76%, the SSE 50 increased by 1.22%, the CSI 300 gained 1.12%, while the CSI 500 and CSI 1000 fell by 0.10% and 0.23% respectively. The ChiNext Index rose by 1.38%, and the Northbound 50 Index increased by 3.13% [2]. - As of May 16, 2025, the valuation percentiles for major indices were categorized as "moderate" for the Shanghai Composite, SSE 50, CSI 300, CSI 500, and CSI 1000, while the ChiNext Index was rated as "safe" [2]. Sector Valuation Analysis - According to the CITIC industry classification, sectors such as non-ferrous metals, electric power and utilities, home appliances, food and beverage, agriculture, non-bank financials, and transportation are rated as "safe" in terms of valuation [2]. Volatility Analysis - The cross-sectional volatility of the CSI 300, CSI 500, and CSI 1000 indices decreased compared to the previous week, indicating a deterioration in the short-term Alpha environment. However, the time series volatility for the CSI 300 increased, suggesting an improvement in the Alpha environment, while the CSI 500 and CSI 1000 saw a decline [3]. Fund Flow Tracking - The top five stocks attracting institutional attention were Anji Technology (241 institutions), Hengerd (237), Naipu Mining Machine (122), Haimu Star (113), and Diaowei (94) [4]. - During the period from May 12 to May 16, 2025, southbound funds experienced a net outflow of 8.685 billion HKD, with net inflows of 4.910 billion HKD in the Shanghai Stock Connect and net outflows of 13.595 billion HKD in the Shenzhen Stock Connect [4]. ETF Performance - The median return for stock ETFs was 0.74% with a net outflow of 25.370 billion CNY. The median return for Hong Kong stock ETFs was 1.27% with a net outflow of 6.696 billion CNY. Cross-border ETFs had a median return of 3.80% with a net outflow of 1.081 billion CNY, while commodity ETFs had a median return of -4.71% with a net outflow of 4.308 billion CNY [6].

查看完整报告 特别申明： 本订阅号中所涉及的证券研究信息由光大证券研究所编写，仅面向光大证券专业投资者客 户，用作新媒体形势下研究信息和研究观点的沟通交流。非光大证券专业投资者客户，请勿 订阅、接收或使用本订阅号中的任何信息。本订阅号难以设置访问权限，若给您造成不便， 敬请谅解。光大证券研究所不会因关注、收到或阅读本订阅号推送内容而视相关人员为光大 证券的客户。 报告摘要 点击注册小程序 本周，基本面因子在多数行业表现较差，其中净利润增长率因子在煤炭行业正收益显著。估值类因子中， BP因子在综合行业正收益显著。流动性因子在交通运输、美容护理、化工、商业贸易和轻工制造行业正收 益显著。市值风格上，本周多数行业表现为小市值风格。 PB-ROE-50组合跟踪： 本周PB-ROE-50组合在中证500、中证800股票池中获取正超额收益。中证500股票池中获得超额收益 0.88%，中证800股票池中获得超额收益0.43%，全市场股票池中获得超额收益-0.02%。 量化市场跟踪 大类因子表现： 本周全市场股票池中，残差波动率因子和盈利因子分别获取正收益0.55%、0.26%；市值因子和非线性市值 因子分别获取负收益-0. ...

- The report tracks the performance of various factors in different stock pools, including the CSI 300, CSI 500, and Liquidity 1500 stock pools[1][2][3] - In the CSI 300 stock pool, the best-performing factors this week were single-quarter net profit year-on-year growth rate (1.02%), single-quarter EPS (1.00%), and PE ratio factor (0.89%)[12][13] - In the CSI 500 stock pool, the best-performing factors this week were EPTTM percentile (1.30%), PB ratio factor (1.07%), and operating cash flow ratio (0.97%)[14][15] - In the Liquidity 1500 stock pool, the best-performing factors this week were post-morning return factor (2.27%), momentum spring factor (1.43%), and PE TTM reciprocal (1.33%)[16][17] - The PB-ROE-50 portfolio achieved positive excess returns in the CSI 500 and CSI 800 stock pools this week, with excess returns of 0.88% and 0.43% respectively[24][25] - The institutional research portfolio tracking strategy achieved positive excess returns this week, with the private equity research tracking strategy achieving an excess return of 0.22% relative to the CSI 800[26][27] - The block trading portfolio achieved a positive excess return of 0.36% relative to the CSI All Share Index this week[30][31] - The directed issuance portfolio achieved a positive excess return of 0.48% relative to the CSI All Share Index this week[35][36]

Quantitative Models and Construction Methods - **Model Name**: PB-ROE-50 **Model Construction Idea**: The model selects stocks based on a combination of Price-to-Book (PB) ratio and Return on Equity (ROE), aiming to capture value and profitability factors[23] **Model Construction Process**: The portfolio is constructed by ranking stocks based on PB and ROE metrics, selecting the top 50 stocks, and rebalancing periodically. Details of the construction process are referenced in earlier reports[23] **Model Evaluation**: The model underperformed this week, delivering negative excess returns across all stock pools, indicating potential challenges in the current market environment[23] - **Model Name**: Public and Private Institution Research Portfolios **Model Construction Idea**: These portfolios are based on stocks that receive significant attention from public and private institutional research, leveraging the informational advantage of institutional focus[25] **Model Construction Process**: Stocks are selected based on the frequency and intensity of institutional research coverage. The portfolios are rebalanced periodically to reflect updated research trends[25] **Model Evaluation**: The public research portfolio achieved positive excess returns this week, while the private research portfolio remained flat, suggesting varying effectiveness of institutional research strategies[25] - **Model Name**: Block Trade Portfolio **Model Construction Idea**: This model identifies stocks with high block trade activity and low volatility, hypothesizing that such stocks exhibit superior subsequent performance[29] **Model Construction Process**: Stocks are ranked based on "block trade transaction ratio" and "6-day transaction volatility." A portfolio is constructed by selecting stocks with high transaction ratios and low volatility, rebalanced monthly[29] **Model Evaluation**: The portfolio delivered strong positive excess returns this week, highlighting the effectiveness of the "high transaction, low volatility" principle[29] - **Model Name**: Private Placement Portfolio **Model Construction Idea**: This model focuses on stocks involved in private placement events, aiming to capture event-driven investment opportunities[34] **Model Construction Process**: Stocks are selected based on private placement announcements, considering factors such as market capitalization, rebalancing frequency, and position control. The portfolio is rebalanced periodically[34] **Model Evaluation**: The portfolio achieved modest positive excess returns this week, indicating the continued relevance of private placement events in generating alpha[34] Model Backtesting Results - **PB-ROE-50 Model**: - Excess return (CSI 500): -0.26% - Excess return (CSI 800): -0.83% - Excess return (All Market): -1.00%[24] - **Public Research Portfolio**: - Excess return (CSI 800): 0.81%[26] - **Private Research Portfolio**: - Excess return (CSI 800): 0.00%[26] - **Block Trade Portfolio**: - Excess return (CSI All Share): 1.55%[30] - **Private Placement Portfolio**: - Excess return (CSI All Share): 0.19%[35] Quantitative Factors and Construction Methods - **Factor Name**: Momentum Factor **Factor Construction Idea**: Captures the momentum effect by identifying stocks with strong recent performance[18] **Factor Construction Process**: Stocks are ranked based on their recent price performance, and the factor is constructed by taking long positions in high-momentum stocks and short positions in low-momentum stocks[18] **Factor Evaluation**: The factor delivered positive returns this week, indicating the presence of momentum effects in the market[18] - **Factor Name**: Nonlinear Market Cap Factor **Factor Construction Idea**: Measures the nonlinear relationship between market capitalization and stock returns[18] **Factor Construction Process**: The factor is derived by fitting a nonlinear regression model to market cap and return data, isolating the nonlinear component[18] **Factor Evaluation**: The factor underperformed this week, reflecting the dominance of small-cap stocks in the market[18] - **Factor Name**: Residual Volatility Factor **Factor Construction Idea**: Identifies stocks with low residual volatility, hypothesizing that such stocks exhibit superior risk-adjusted returns[18] **Factor Construction Process**: Residual volatility is calculated by regressing stock returns on market returns and measuring the standard deviation of residuals. Stocks with low residual volatility are favored[18] **Factor Evaluation**: The factor delivered negative returns this week, suggesting a challenging environment for low-volatility strategies[18] Factor Backtesting Results - **Momentum Factor**: Weekly return: 0.69%[18] - **Nonlinear Market Cap Factor**: Weekly return: -0.58%[18] - **Residual Volatility Factor**: Weekly return: -0.64%[18]

Group 1 - The core viewpoint of the article highlights the performance of various market factors, indicating that the leverage factor achieved a positive return of 0.60%, while liquidity and market capitalization factors recorded negative returns of -0.77% and -0.52% respectively, suggesting a significant small-cap style in the market [2] Group 2 - In the CSI 300 stock pool, the best-performing factors included momentum-adjusted large orders (2.40%), net inflow of large orders (2.32%), and momentum-adjusted small orders (1.82%). Conversely, the worst-performing factors were gross profit margin on total assets for the quarter (-2.60%), total asset growth rate (-2.48%), and return on assets for the quarter (-2.14%) [3] - In the CSI 500 stock pool, the top-performing factors were momentum-adjusted small orders (2.99%), price-to-earnings ratio factor (2.89%), and price-to-book ratio factor (2.75%). The underperforming factors included the standard deviation of 5-day trading volume (-0.57%), TTM net profit margin (-0.48%), and TTM operating profit margin (-0.27%) [3] - In the liquidity 1500 stock pool, the leading factors were price-to-earnings ratio factor (2.79%), inverse of TTM price-to-earnings ratio (2.10%), and correlation between intraday volatility and trading volume (1.89%). The lagging factors were gross profit margin on total assets for the quarter (-2.31%), standardized unexpected income (-2.15%), and standardized unexpected profit (-2.06%) [3] Group 3 - The performance of fundamental factors varied across industries, with net asset growth rate, net asset per share, and TTM operating profit per share factors showing consistent performance in the environmental protection industry. Valuation factors such as BP and EP factors performed consistently in the home appliances, machinery, environmental protection, building materials, and chemical industries. Residual volatility and liquidity factors yielded consistent positive returns in the food and beverage, non-ferrous metals, oil and petrochemicals, real estate, and comprehensive industries. The small-cap style was notably significant across most industries this week [4] Group 4 - The PB-ROE-50 combination experienced a pullback in excess returns across various stock pools, with the CSI 500 pool recording an excess return of -1.20%, the CSI 800 pool at -2.24%, and the overall market stock pool at -2.22% [5] Group 5 - This week, the public fund research stock selection strategy and private fund research tracking strategy achieved positive excess returns, with the public fund strategy gaining 0.33% relative to the CSI 800 and the private fund strategy gaining 0.51% [8] Group 6 - The block trading combination achieved a positive excess return of 0.78% relative to the CSI All Share Index this week [9] Group 7 - The targeted issuance combination gained a significant excess return of 3.07% relative to the CSI All Share Index this week [10]

Group 1 - The core viewpoint of the article emphasizes that the spring market trend is expected to continue, driven by policy and economic data catalysts, with a focus on growth and consumer sectors [4] - The A-share market has experienced significant volatility, with major indices declining, particularly the ChiNext Index, while growth stocks and small-cap stocks are expected to outperform [5] - The oil and chemical sectors are poised for recovery due to easing geopolitical tensions, benefiting downstream refining companies from reduced cost pressures [7][8] Group 2 - The low-altitude economy is projected to reach a market size of 1.5 trillion yuan in 2025, with significant growth in the humanoid robot market, expected to grow from approximately 1.19 million units to 60.57 million units by 2030 [9] - The Hong Kong stock market has seen increased trading activity since September 2024, leading to record financial performance, with anticipated boosts from mainland China's stimulus policies [10]