Quantitative Models and Construction 1. Model Name: GRU Model - **Model Construction Idea**: The GRU model is used to predict stock returns based on historical data and incorporates various factors to optimize portfolio performance [3][4][5] - **Model Construction Process**: - The GRU model is trained on historical data to capture temporal dependencies in stock returns - It uses multiple input features, including technical and fundamental factors, to predict future returns - The model is applied to different stock pools (e.g., CSI 300, CSI 500, CSI 1000) to evaluate its performance [5][6][7] - **Model Evaluation**: The GRU model demonstrates strong performance in most stock pools, with positive long-short returns across various factors. However, certain sub-models (e.g., `barra5d`) show occasional underperformance [5][6][7] 2. Model Name: Open1d and Close1d Models - **Model Construction Idea**: These models focus on short-term price movements and are designed to capture daily return patterns [8][31] - **Model Construction Process**: - Open1d and Close1d models are trained on daily open and close price data, respectively - They are evaluated based on their ability to generate excess returns relative to the CSI 1000 index [8][31] - **Model Evaluation**: These models show mixed performance, with occasional drawdowns relative to the benchmark index [8][31] 3. Model Name: Barra1d and Barra5d Models - **Model Construction Idea**: These models are based on the Barra factor framework and aim to capture short-term and medium-term return patterns [8][31] - **Model Construction Process**: - Barra1d focuses on daily factor returns, while Barra5d aggregates returns over a 5-day horizon - Both models are tested for their ability to generate excess returns relative to the CSI 1000 index [8][31] - **Model Evaluation**: Barra5d demonstrates strong year-to-date performance, significantly outperforming the benchmark, while Barra1d shows consistent but less pronounced gains [8][31] --- Model Backtest Results 1. GRU Model - **Excess Return**: Positive across most stock pools, with occasional underperformance in specific sub-models like `barra5d` [5][6][7] 2. Open1d Model - **Weekly Excess Return**: -0.01% - **Year-to-Date Excess Return**: 5.23% [32] 3. Close1d Model - **Weekly Excess Return**: -0.38% - **Year-to-Date Excess Return**: 3.64% [32] 4. Barra1d Model - **Weekly Excess Return**: 0.65% - **Year-to-Date Excess Return**: 3.80% [32] 5. Barra5d Model - **Weekly Excess Return**: 0.02% - **Year-to-Date Excess Return**: 6.44% [32] --- Quantitative Factors and Construction 1. Factor Name: Beta - **Factor Construction Idea**: Measures historical beta to capture market sensitivity [15] - **Factor Construction Process**: Historical beta is calculated based on the covariance of stock returns with market returns [15] 2. Factor Name: Momentum - **Factor Construction Idea**: Captures historical excess return trends [15] - **Factor Construction Process**: - Momentum = 0.74 * Historical Excess Return Volatility + 0.16 * Cumulative Excess Return Deviation + 0.1 * Historical Residual Return Volatility [15] 3. Factor Name: Volatility - **Factor Construction Idea**: Measures stock price fluctuations to identify high-volatility stocks [15] - **Factor Construction Process**: - Volatility = Weighted combination of historical residual return volatility and other metrics [15] 4. Factor Name: Growth - **Factor Construction Idea**: Focuses on earnings and revenue growth rates [15] - **Factor Construction Process**: - Growth = 0.24 * Earnings Growth Rate + 0.47 * Revenue Growth Rate [15] 5. Factor Name: Liquidity - **Factor Construction Idea**: Measures stock turnover to identify liquid stocks [15] - **Factor Construction Process**: - Liquidity = 0.35 * Monthly Turnover + 0.35 * Quarterly Turnover + 0.3 * Annual Turnover [15] --- Factor Backtest Results 1. Beta Factor - **Weekly Long-Short Return**: Positive [16][18] 2. Momentum Factor - **Weekly Long-Short Return**: Negative [16][18] 3. Volatility Factor - **Weekly Long-Short Return**: Positive [16][18] 4. Growth Factor - **Weekly Long-Short Return**: Positive [16][18] 5. Liquidity Factor - **Weekly Long-Short Return**: Positive [16][18]

Quantitative Models and Construction Methods 1. Model Name: Layered Progressive Stock Selection for Profitability Factor - **Model Construction Idea**: The model aims to enhance the profitability factor by progressively filtering stocks based on historical ROE and financial stability, ensuring higher future ROE probabilities [38][35][36] - **Model Construction Process**: - Step 1: Select the top 100 stocks based on historical ROE (ROE_ttm) [38] - Step 2: From the top 100, further filter the top 50 stocks with the highest financial stability scores, which include metrics like ROE stability, revenue growth stability, and leverage stability [27][38] - Step 3: Construct an equal-weighted portfolio with the final 50 stocks [38] - **Model Evaluation**: The layered approach effectively reduces the probability of ROE decline by one interval (5%) and increases the likelihood of maintaining high ROE levels in the future [38][36] 2. Model Name: Dividend Growth Factorization - **Model Construction Idea**: This model predicts future dividend growth by constructing a stock pool based on historical dividend stability and earnings growth expectations [49][51] - **Model Construction Process**: - Step 1: Select stocks with stable dividend payout ratios over the past three years and positive earnings growth expectations [49] - Step 2: Select stocks with dividend amounts growing over the past two years and positive earnings growth expectations [49] - Step 3: Combine the two pools to form a comprehensive stock pool [49] - Step 4: Construct sub-factors such as dividend payout deviation, sell-side forecast count, and recent financial report growth, standardize and sum them, and take the maximum value across perspectives [51] - **Model Evaluation**: The model improves the prediction accuracy of dividend growth, achieving over a 10% improvement in win rates for both the CSI 300 and CSI 500 indices [51][52] 3. Model Name: Growth Factor Improvement via Reverse Exclusion - **Model Construction Idea**: Instead of further refining high-growth stocks, this model excludes stocks unlikely to achieve future net profit growth, enhancing the growth factor's predictive power [70][69] - **Model Construction Process**: - Step 1: Start with 100 high-growth stocks based on historical growth factors [70] - Step 2: Exclude stocks meeting any of the following conditions: - FY1 consensus forecast ≤ 0 - FY1 consensus forecast is null - Consensus forecast downgraded in the past 4, 13, or 26 weeks [70] - Step 3: Construct a portfolio with the remaining stocks [70] - **Model Evaluation**: The exclusion method significantly improves the prediction rate of actual net profit growth and reduces the probability of selecting companies with declining net profits [70][69] 4. Model Name: Composite Three-Factor Portfolio - **Model Construction Idea**: This model integrates the improved profitability, dividend, and growth factors into a unified portfolio to enhance index performance [81][83] - **Model Construction Process**: - Step 1: Combine the stock pools from the three improved factors (profitability, dividend, growth) [81] - Step 2: Select approximately 120 stocks from the combined pool, ensuring industry neutrality and periodic rebalancing [83] - **Model Evaluation**: The composite portfolio demonstrates consistent performance improvement over the equal-weighted three-factor portfolio, with notable gains in the CSI 300 and CSI 500 indices [83][86] 5. Model Name: Three-Factor Portfolio + Volume-Price Factors - **Model Construction Idea**: This model incorporates volume-price factors (low volatility, low liquidity, momentum) into the three-factor portfolio to capture additional returns during strong volume-price factor periods [100][97] - **Model Construction Process**: - Step 1: Start with the three-factor composite portfolio [100] - Step 2: Select the top 75 stocks based on volume-price factor scores (low volatility, low liquidity, momentum) [100] - Step 3: Construct an equal-weighted portfolio with the selected stocks [100] - **Model Evaluation**: The addition of volume-price factors further enhances long-term returns and maintains stable excess returns compared to the equal-weighted six-factor portfolio [100][103] 6. Model Name: 75+25 Composite Portfolio - **Model Construction Idea**: This model combines the three-factor portfolio with a 25-stock pool selected based on volume-price factors across the entire market, aiming to maximize expected returns [109][112] - **Model Construction Process**: - Step 1: Select 75 stocks from the three-factor portfolio [109] - Step 2: Select 25 stocks from the entire market based on volume-price factors (growth, profitability, low volatility, small market cap) [109] - Step 3: Combine the two pools into a 100-stock portfolio [109] - **Model Evaluation**: The 75+25 portfolio achieves significant improvements in annualized returns and Sharpe ratios, benefiting from the strong performance of volume-price factors in recent years [112][125] --- Model Backtest Results 1. Layered Progressive Stock Selection for Profitability Factor - CSI 300: Win rate improved from 78.03% to 86.28% [36] - CSI 500: Win rate improved from 78.72% to 86.55% [36] 2. Dividend Growth Factorization - CSI 300: Win rate improved from 54.90% to 73.24% [51] - CSI 500: Win rate improved from 40.14% to 54.28% [51] 3. Growth Factor Improvement via Reverse Exclusion - CSI 300: Win rate improved from 83.38% to 92.88% [69] - CSI 500: Win rate improved from 80.21% to 90.13% [69] 4. Composite Three-Factor Portfolio - CSI 300: Annualized return improved from 6.36% to 9.34%, Sharpe ratio improved from 0.34 to 0.49 [86] - CSI 500: Annualized return improved from 5.46% to 7.36%, Sharpe ratio improved from 0.26 to 0.34 [86] 5. Three-Factor Portfolio + Volume-Price Factors - CSI 300: Annualized return improved from 7.81% to 11.55%, Sharpe ratio improved from 0.40 to 0.62 [103] - CSI 500: Annualized return improved from 6.75% to 9.15%, Sharpe ratio improved from 0.32 to 0.45 [103] 6. 75+25 Composite Portfolio - CSI 300: Annualized return improved from 7.84% to 14.56%, Sharpe ratio improved from 0.41 to 0.75 [112] - CSI 500: Annualized return improved from 7.35% to 13.18%, Sharpe ratio improved from 0.36 to 0.62 [112]

Group 1: Report Industry Investment Rating - No relevant content provided Group 2: Report's Core View - The commodity market shows different trends in various sectors. The bearish proportion has slightly increased this week, with significant changes in the black and agricultural sectors. The overall market signals are mainly bearish, with some exceptions like the iron ore market turning neutral [1]. Group 3: Summary by Related Content Commodity Market Sector Analysis - The agricultural sector's momentum is rising, while the energy sector is relatively weak. In the black sector, the momentum factor has decreased marginally, and the term - structure differentiation has narrowed. In the non - ferrous sector, the position - holding factor has decreased marginally, and the cross - sectional differentiation has widened. In the energy - chemical sector, there is cross - sectional momentum differentiation. In the agricultural sector, the position - holding of oilseeds and meals has increased, and the short - cycle momentum of palm oil has rebounded [1]. Performance of Different Commodities Methanol - Last week, the supply factor strengthened by 0.64%, the inventory factor declined by 0.66%, and the comprehensive signal this week is bearish. Fundamentally, the supply side remains bearish, the demand side is neutral to bearish, the inventory side turns bearish, and the spread side is neutral to bearish [1]. Float Glass - Last week, the inventory factor increased by 2.47%, the spread factor weakened by 0.17%, the profit factor decreased by 0.20%, and the synthetic factor declined by 0.13%, with a bearish comprehensive signal this week. Fundamentally, the supply side is neutral, the demand side is neutral to bearish, the inventory side is bearish, and the profit side is neutral to bullish [1]. Iron Ore - Last week, the supply factor strengthened by 0.38%, the synthetic factor increased by 0.08%, and the comprehensive signal this week turns neutral. Fundamentally, the supply side's bearish feedback weakens to neutral, the demand side turns to bullish feedback but remains neutral, the inventory side turns bearish, and the spread side turns bullish [3][4].

- The report tracks high-frequency stock selection factors, including Price Range Factor, Price-Volume Divergence Factor, Regret Avoidance Factor, and Slope Convexity Factor, with their out-of-sample performance showing overall excellence[2][3][11] - Price Range Factor measures the activity level of stocks traded within different intraday price ranges, reflecting investors' expectations for future stock trends. It demonstrates strong predictive power and stable performance this year[3][11][17] - Price-Volume Divergence Factor evaluates the correlation between stock prices and trading volumes. Lower correlation typically indicates higher potential for future stock price increases. However, its performance has been unstable in recent years, with multi-long net value curves flattening[3][22][26] - Regret Avoidance Factor examines the proportion and degree of stock rebounds after being sold by investors, showcasing good predictive power. Its out-of-sample excess returns are stable, indicating that A-share investors' regret avoidance sentiment significantly impacts stock price expectations[3][27][36] - Slope Convexity Factor analyzes the slope and convexity of order books to assess the impact of investor patience and supply-demand elasticity on expected returns. It is constructed using high-frequency snapshot data from limit order books[3][37][42] - The report combines three high-frequency factors into an equal-weighted "Gold" portfolio for CSI 1000 Index enhancement strategy, achieving an annualized excess return rate of 10.51% and a maximum excess drawdown of 6.04%[3][44][45] - To further enhance strategy performance, the report integrates high-frequency factors with three effective fundamental factors (Consensus Expectations, Growth, and Technical Factors) to construct a high-frequency & fundamental resonance portfolio for CSI 1000 Index enhancement strategy. This strategy achieves an annualized excess return rate of 14.57% and a maximum excess drawdown of 4.52%[4][49][51] Factor Backtesting Results - Price Range Factor: Weekly excess return 0.40%, monthly excess return 0.51%, annual excess return 5.86%[2][13][17] - Price-Volume Divergence Factor: Weekly excess return -0.24%, monthly excess return 1.53%, annual excess return 9.00%[2][13][26] - Regret Avoidance Factor: Weekly excess return 0.27%, monthly excess return -0.49%, annual excess return 2.32%[2][13][36] - Slope Convexity Factor: Weekly excess return -1.74%, monthly excess return -2.46%, annual excess return -5.90%[2][13][42] Strategy Performance Metrics - "Gold" Portfolio: Annualized return 9.49%, annualized excess return 10.51%, Sharpe ratio 0.39, IR 2.47, maximum excess drawdown 6.04%[45][47][48] - High-frequency & Fundamental Resonance Portfolio: Annualized return 13.62%, annualized excess return 14.57%, Sharpe ratio 0.58, IR 3.50, maximum excess drawdown 4.52%[51][53][55]

Group 1 - The core viewpoint is that quantitative strategies in investment rely on diverse data sources, including traditional financial reports and non-traditional data such as social media sentiment and supply chain information, to generate forward-looking investment signals [1][2] - The performance of growth factors has been relatively strong in the current year, indicating a favorable market environment for growth-oriented investments [1] - A rich factor library is essential for diversifying sources of returns and enhancing cyclical resilience, necessitating regular updates to the factor pool to include both economically supported and algorithmically derived factors [1] Group 2 - Continuous model iteration and an open attitude towards new technologies, particularly AI, are crucial for improving the efficiency of factor development and constructing stronger predictive signals [2] - The application of AI in quantitative investment processes has become increasingly prevalent, including the use of large models for text data analysis and advanced models like transformers for end-to-end factor mining [2]

Group 1: Overall Market Conditions - The proportion of short positions in commodities increased this week, mainly due to the decline in the factor strength of the energy and chemical sector. Currently, the relatively strong sectors in the cross - section are precious metals and agricultural products, while the relatively weak sector is the energy sector [3]. - In the precious metals sector, the sequential momentum of gold has marginally recovered, and the differentiation within the sector has narrowed. In the non - ferrous sector, the position factor has continued to decline slightly, and the cross - sectional differentiation has expanded, with copper and zinc being on the weaker end [3]. - In the black sector, the short - term momentum factor has marginally decreased, but the long - term factor has gradually stabilized, and the term structure differentiation has narrowed. In the energy and chemical sector, the cross - sectional momentum has declined overall, and PTA, soda ash, and glass are on the weaker end of the sector's cross - section [3]. - In the agricultural products sector, the positions of oilseeds and meals have both decreased slightly, and the differentiation has narrowed [3]. Group 2: Factor Returns - Last week's returns and monthly returns for different factors: supply factor had a last - week return of 1.64% and a monthly return of 0.00%; demand factor had a last - week return of 1.51% and a monthly return of 0.00%; inventory factor had a last - week return of 1.20% and a monthly return of - 2.28%; spread factor had a last - week return of 3.90% and a monthly return of 2.50%; profit factor had a last - week return of 0.00% and a monthly return of 0.00%; the cumulative return of major categories last week was 1.64% and this month was - 0.04% [7]. Group 3: Methanol Analysis - In terms of strategy net value, last week, the supply factor strengthened by 0.21%, the demand factor increased by 0.13%, the inventory factor weakened by 0.19%, the spread factor increased by 0.09%, and the composite factor increased by 0.16%. This week, the comprehensive signal for short positions continues. On the fundamental factor side, the capacity utilization rate of domestic methanol plants has increased, and the import arrival volume has increased slightly, indicating a short position on the supply side; the operating rates of formaldehyde and chloride plants have increased, but the capacity utilization rates of acetic acid and MTBE plants have decreased, making the demand side neutral; the inventory of domestic methanol manufacturers has decreased, indicating a long position on the inventory side; the closing price of the main methanol futures contract and the 9 - 1 spread on the futures market have both released short - position signals, and the spread side has turned to a short position [4]. Group 4: Glass Analysis - In terms of strategy net value, last week, the inventory factor decreased by 2.28%, the spread factor increased by 2.50%, and the composite factor weakened by 0.04%. This week, the comprehensive signal is a short position. On the fundamental factor side, the capacity utilization rate of float glass has remained flat month - on - month, keeping the supply side neutral; the number of commercial housing transactions in 30 large - and medium - sized Chinese cities has increased slightly, making the demand side neutral; float glass enterprises have continued to reduce inventory, indicating a long position on the inventory side; the spot prices of the float glass markets in Central China, North China, and South China have all released short - position signals, indicating a short position on the spread side; the pre - tax gross profit of float glass made from steam coal has declined, indicating a short position on the profit side [7]. Group 5: Iron Ore Analysis - In terms of strategy net value, last week, each factor remained unchanged, and this week, the comprehensive signal remains neutral. The arrival volume at Qingdao Port has increased significantly, and the shipping volumes of BHP and Rio Tinto have increased, turning the supply side into a short - position feedback, but the overall signal remains neutral. The consumption of iron ore powder for sintering in steel mills and the proportion of sintered ore in the furnace have decreased, and the strength of the long - position feedback on the demand side has decreased slightly, but the signal is still neutral. The iron ore concentrate at ports and the domestic sintering iron ore powder in steel mills have both reduced inventory slightly, weakening the short - position feedback on the inventory side. The price center of PB powder has shifted down, further weakening the strength of the long - position feedback on the spread side [7]. Group 6: Lead Analysis - In terms of strategy net value, last week, the supply factor strengthened by 0.52%, the demand factor decreased by 0.51%, the spread factor increased by 0.46%, and the composite factor strengthened by 0.15%. This week, the comprehensive signal has changed from a short position to neutral. The price of domestic lead concentrates from SMM has decreased, and the profit of tax - free recycled lead has decreased, turning the supply - side signal to neutral. The LME lead inventory and registered warrants have increased, turning the inventory side into a short - position feedback, but the overall signal remains neutral. The spread between the near and far months of LME lead has widened, strengthening the short - position feedback on the spread side [7].

Quantitative Models and Construction Methods - **Model Name**: Value Smart Beta Portfolio **Model Construction Idea**: The Value Smart Beta portfolios are constructed based on the goal of achieving high beta elasticity and long-term stable excess returns. The portfolios are designed to capture the value style with low historical correlation to other styles[7] **Model Construction Process**: Two portfolios are constructed under the value style: the "Value 50 Portfolio" and the "Value Balanced 50 Portfolio". These portfolios are selected and weighted to optimize for the value factor while maintaining diversification and minimizing correlation with other factors[7] **Model Evaluation**: The Value Smart Beta portfolios demonstrate a focus on capturing value-oriented excess returns, but their performance is sensitive to market conditions[7] - **Model Name**: Growth Smart Beta Portfolio **Model Construction Idea**: The Growth Smart Beta portfolios aim to capture the growth style with high beta elasticity and long-term stable excess returns. These portfolios are designed to have low historical correlation with other styles[7] **Model Construction Process**: Two portfolios are constructed under the growth style: the "Growth 50 Portfolio" and the "Growth Balanced 50 Portfolio". The portfolios are optimized to emphasize growth characteristics while maintaining diversification[7] **Model Evaluation**: The Growth Smart Beta portfolios are effective in capturing growth-oriented returns but may underperform in value-dominated market conditions[7] - **Model Name**: Small-Cap Smart Beta Portfolio **Model Construction Idea**: The Small-Cap Smart Beta portfolios are designed to capture the small-cap style with high beta elasticity and long-term stable excess returns. These portfolios are constructed to have low historical correlation with other styles[7] **Model Construction Process**: Two portfolios are constructed under the small-cap style: the "Small-Cap 50 Portfolio" and the "Small-Cap Balanced 50 Portfolio". The portfolios are optimized to emphasize small-cap characteristics while maintaining diversification[7] **Model Evaluation**: The Small-Cap Smart Beta portfolios show strong performance in capturing small-cap excess returns, particularly in favorable market environments[7] --- Model Backtesting Results - **Value 50 Portfolio**: - Weekly Absolute Return: -2.12% - Weekly Excess Return: -0.41% - Monthly Absolute Return: 0.20% - Monthly Excess Return: 0.33% - Year-to-Date Absolute Return: 12.44% - Year-to-Date Excess Return: 8.78% - Maximum Relative Drawdown: 2.35%[8] - **Value Balanced 50 Portfolio**: - Weekly Absolute Return: -0.46% - Weekly Excess Return: 1.26% - Monthly Absolute Return: 0.48% - Monthly Excess Return: 0.61% - Year-to-Date Absolute Return: 10.16% - Year-to-Date Excess Return: 6.50% - Maximum Relative Drawdown: 3.99%[8] - **Growth 50 Portfolio**: - Weekly Absolute Return: -1.48% - Weekly Excess Return: 0.68% - Monthly Absolute Return: -0.71% - Monthly Excess Return: -0.31% - Year-to-Date Absolute Return: 4.50% - Year-to-Date Excess Return: 2.38% - Maximum Relative Drawdown: 3.61%[8] - **Growth Balanced 50 Portfolio**: - Weekly Absolute Return: -1.64% - Weekly Excess Return: 0.53% - Monthly Absolute Return: 0.06% - Monthly Excess Return: 0.46% - Year-to-Date Absolute Return: 8.71% - Year-to-Date Excess Return: 6.59% - Maximum Relative Drawdown: 6.11%[8] - **Small-Cap 50 Portfolio**: - Weekly Absolute Return: 1.25% - Weekly Excess Return: 1.43% - Monthly Absolute Return: 1.07% - Monthly Excess Return: 0.85% - Year-to-Date Absolute Return: 36.52% - Year-to-Date Excess Return: 19.90% - Maximum Relative Drawdown: 6.23%[8] - **Small-Cap Balanced 50 Portfolio**: - Weekly Absolute Return: -1.09% - Weekly Excess Return: -0.90% - Monthly Absolute Return: 0.61% - Monthly Excess Return: 0.39% - Year-to-Date Absolute Return: 26.60% - Year-to-Date Excess Return: 9.98% - Maximum Relative Drawdown: 4.56%[8]

- The report tracks the performance of style factors, including Beta, liquidity, leverage, profitability, and market capitalization, with Beta and liquidity showing strong long positions, while leverage, profitability, and market capitalization exhibit strong short positions [2][16] - Style factors are constructed using various metrics, such as historical Beta, logarithm of total market capitalization, historical excess return averages for momentum, and a weighted combination of volatility measures for the volatility factor. For example, the volatility factor is calculated as $ 0.74 * historical excess return volatility + 0.16 * cumulative excess return deviation + 0.1 * historical residual return volatility $ [15] - Fundamental factors, including growth-related financial metrics and static financial metrics, are tested across different stock pools (e.g., CSI 300, CSI 500, CSI 1000). Growth-related financial factors generally show mixed or negative performance, while static financial factors exhibit varied results depending on the stock pool [3][4][5][6][18][20][23][25] - Technical factors, such as momentum and volatility, generally show positive performance across stock pools, with high-volatility and high-momentum stocks being dominant. For example, the 120-day momentum factor and 20-day volatility factor are highlighted for their significant contributions [3][4][5][6][18][20][23][26] - GRU factors are tested using different models (e.g., barra1d, barra5d, close1d), with performance varying across stock pools. For instance, barra1d shows strong positive performance in CSI 500 and CSI 1000 pools, while close1d experiences significant drawdowns in CSI 1000 [3][4][5][6][18][20][23][26] - Multi-factor strategies and GRU-based long portfolios are evaluated against the CSI 1000 index. GRU long portfolios show weak performance this week, with relative drawdowns of 0.11%-0.25%, while the barra5d model demonstrates strong year-to-date performance, achieving an excess return of 8.36% [7][30][31]

智通财经记者 | 龙力 量化私募开始"卷调研"了？ 私募排排网数据显示，截至7月24日，今年以来一共有137家量化私募累计调研达581次，涉及到29个申 万一级行业中的408只个股。其中，有15家量化私募今年以来的调研总次数不少于10次。 | 表 | | --- | 格：今年以来调研总次数不少于10次的量化私募名单（截至7月24日）数据来源：私募排排网、智通财 经整理 排排网集团旗下融智投资FOF基金经理李春瑜对智通财经表示，量化私募机构对上市公司的调研活动明 显升温，这一现象主要受两大因素推动：其一，部分头部量化机构正着手组建主观投研团队，以丰富其 多策略投资体系，增强收益的多元化；其二，在行业管理规模快速扩张的背景下，传统量价因子的超额 收益持续衰减，促使量化私募转向更深入的基本面研究，而上市公司调研正是获取高质量基本面数据的 关键环节。 "量化私募调研升温的核心动因是策略收益来源的多元化需求和市场环境与监管驱动的范式转型。" 尚 艺投资总经理王峥对智通财经表示。 王峥进一步对智通财经解释称，首先传统量价因子因策略同质化与市场有效性提升，超额收益持续衰 减，量化私募可以通过组建主观研究团队，将基本面分析融 ...

Quantitative Models and Construction Methods - **Model Name**: GRU **Model Construction Idea**: GRU is used for industry rotation and stock selection based on historical data and market trends[3][5][7] **Model Construction Process**: GRU utilizes gated recurrent units to process sequential data, capturing temporal dependencies in stock price movements and industry performance. It incorporates multiple factors such as momentum, volatility, and valuation metrics to predict future trends[3][5][7] **Model Evaluation**: GRU demonstrates strong performance in multi-factor combinations and industry rotation strategies, with notable differentiation across different stock pools[3][5][7] - **Model Name**: Barra **Model Construction Idea**: Barra focuses on style factors to explain stock returns and risks[14][15][16] **Model Construction Process**: Barra includes multiple style factors such as Beta, Size, Momentum, Volatility, Non-linear Size, Valuation, Liquidity, Profitability, Growth, and Leverage. Each factor is calculated using specific formulas: - **Beta**: Historical beta - **Size**: Natural logarithm of total market capitalization - **Momentum**: Mean of historical excess return series - **Volatility**: $0.74 \times \text{historical excess return volatility} + 0.16 \times \text{cumulative excess return deviation} + 0.1 \times \text{historical residual return volatility}$ - **Non-linear Size**: Cubic transformation of market capitalization - **Valuation**: Reciprocal of price-to-book ratio - **Liquidity**: $0.35 \times \text{monthly turnover rate} + 0.35 \times \text{quarterly turnover rate} + 0.3 \times \text{annual turnover rate}$ - **Profitability**: Weighted combination of analyst forecast earnings-price ratio, reciprocal of cash flow ratio, reciprocal of trailing twelve-month P/E ratio, and forecasted growth rates - **Growth**: Weighted combination of earnings growth rate and revenue growth rate - **Leverage**: Weighted combination of market leverage, book leverage, and debt-to-asset ratio[15] **Model Evaluation**: Barra style factors provide a comprehensive framework for analyzing stock returns, with strong differentiation in multi-factor strategies[14][15][16] Model Backtesting Results - **GRU Model**: - **open1d**: Weekly excess return 0.61%, monthly 1.56%, yearly 7.78% - **close1d**: Weekly excess return 0.02%, monthly 1.45%, yearly 7.28% - **barra1d**: Weekly excess return -0.24%, monthly -0.07%, yearly 3.61% - **barra5d**: Weekly excess return 0.06%, monthly 1.35%, yearly 8.63% - **Multi-factor combination**: Weekly excess return 0.61%, monthly 0.82%, yearly 3.22%[31][32][33] Quantitative Factors and Construction Methods - **Factor Name**: Beta **Factor Construction Idea**: Measures historical sensitivity to market movements[15] **Factor Construction Process**: Calculated as historical beta using regression analysis of stock returns against market returns[15] - **Factor Name**: Size **Factor Construction Idea**: Captures the impact of market capitalization on stock returns[15] **Factor Construction Process**: Natural logarithm of total market capitalization[15] - **Factor Name**: Momentum **Factor Construction Idea**: Reflects the persistence of stock price trends[15] **Factor Construction Process**: Mean of historical excess return series[15] - **Factor Name**: Volatility **Factor Construction Idea**: Measures risk associated with stock price fluctuations[15] **Factor Construction Process**: $0.74 \times \text{historical excess return volatility} + 0.16 \times \text{cumulative excess return deviation} + 0.1 \times \text{historical residual return volatility}$[15] - **Factor Name**: Non-linear Size **Factor Construction Idea**: Captures non-linear effects of market capitalization on returns[15] **Factor Construction Process**: Cubic transformation of market capitalization[15] - **Factor Name**: Valuation **Factor Construction Idea**: Reflects the relative attractiveness of stock prices[15] **Factor Construction Process**: Reciprocal of price-to-book ratio[15] - **Factor Name**: Liquidity **Factor Construction Idea**: Measures ease of trading stocks[15] **Factor Construction Process**: $0.35 \times \text{monthly turnover rate} + 0.35 \times \text{quarterly turnover rate} + 0.3 \times \text{annual turnover rate}$[15] - **Factor Name**: Profitability **Factor Construction Idea**: Captures earnings quality and growth potential[15] **Factor Construction Process**: Weighted combination of analyst forecast earnings-price ratio, reciprocal of cash flow ratio, reciprocal of trailing twelve-month P/E ratio, and forecasted growth rates[15] - **Factor Name**: Growth **Factor Construction Idea**: Reflects revenue and earnings growth trends[15] **Factor Construction Process**: Weighted combination of earnings growth rate and revenue growth rate[15] - **Factor Name**: Leverage **Factor Construction Idea**: Measures financial risk associated with debt levels[15] **Factor Construction Process**: Weighted combination of market leverage, book leverage, and debt-to-asset ratio[15] Factor Backtesting Results - **Beta**: Weekly excess return -0.24%, monthly -0.07%, yearly 3.61%[31][32][33] - **Size**: Weekly excess return 0.02%, monthly 1.45%, yearly 7.28%[31][32][33] - **Momentum**: Weekly excess return 0.61%, monthly 1.56%, yearly 7.78%[31][32][33] - **Volatility**: Weekly excess return 0.06%, monthly 1.35%, yearly 8.63%[31][32][33] - **Non-linear Size**: Weekly excess return 0.61%, monthly 0.82%, yearly 3.22%[31][32][33] - **Valuation**: Weekly excess return 0.61%, monthly 0.82%, yearly 3.22%[31][32][33] - **Liquidity**: Weekly excess return 0.61%, monthly 0.82%, yearly 3.22%[31][32][33] - **Profitability**: Weekly excess return 0.61%, monthly 0.82%, yearly 3.22%[31][32][33] - **Growth**: Weekly excess return 0.61%, monthly 0.82%, yearly 3.22%[31][32][33] - **Leverage**: Weekly excess return 0.61%, monthly 0.82%, yearly 3.22%[31][32][33]