Investment Rating - The investment rating for the company is "Buy" [2] Core Views - The company is expected to achieve profitability in 2025, being the first among major airlines to turn a profit in Q2 2025. The overall demand fluctuations do not alter the long-term growth logic of the aviation industry, and an optimistic upward trend in profitability is anticipated over the next two years [3][11] - The company has maintained a target price of 13.52 CNY for 2027, based on a projected PE ratio of 16 times [11] Financial Summary - Total revenue is projected to grow from 141.1 billion CNY in 2023 to 204.739 billion CNY by 2027, reflecting a compound annual growth rate (CAGR) of approximately 10.3% [5][12] - The net profit attributable to the parent company is expected to shift from a loss of 1.046 billion CNY in 2023 to a profit of 15.07 billion CNY in 2027, indicating a significant recovery and growth trajectory [5][12] - Earnings per share (EPS) is forecasted to improve from -0.06 CNY in 2023 to 0.86 CNY in 2027 [5][12] - The return on equity (ROE) is projected to rise from -2.8% in 2023 to 22.6% in 2027, showcasing a strong recovery in profitability [5][12] Strategic Developments - The company is set to maintain its controlling stake in Shenzhen Airlines, which is undergoing a capital increase of 16 billion CNY. This move is expected to alleviate debt burdens and enhance profitability [11] - The company’s network and customer quality are continuously improving, positioning it as a leading player in the aviation sector [11] Market Position - The company has a total market capitalization of 135.4 billion CNY, with a current stock price of 7.76 CNY [6][12] - The company’s stock has shown resilience, with a 52-week price range of 6.85 to 8.89 CNY [6]

Investment Rating - The report does not explicitly state an investment rating for the industry Core Insights - The embodied intelligence industry is primarily represented by embodied robots and smart vehicles, which are mutually reinforcing in their development [7] - Recent advancements include the unveiling of Fourier's third-generation humanoid robot GR-3C and the establishment of the largest humanoid robot training center in China [10][11] - Significant investment activities have been observed in the embodied intelligence sector, with multiple companies securing funding to enhance their technological capabilities and market presence [24] Summary by Sections 1. Embodied Robot and Smart Vehicle Dynamics - Fourier's third-generation humanoid robot GR-3C was showcased at the Industrial Expo, featuring 55 degrees of freedom and advanced interaction capabilities [10] - A new humanoid robot training center in Beijing can produce over 6 million data points annually, enhancing the practical application of humanoid robots [11] - Cross-dimensional intelligence released an upgraded video learning framework that allows rapid task adaptation across different robot bodies [12] - DeepMind introduced the Gemini Robotics 1.5 series, enhancing robots' capabilities in visual and language processing [13] - A strategic partnership between Lingyi Robotics and AIRS aims to integrate hardware and algorithms for humanoid robot development [15] - Kepler has initiated mass production of the world's first commercially available hybrid humanoid robot, K2 "Bumblebee," with significant pre-orders [16] 2. Smart Vehicle Dynamics - The sales ranking for new energy vehicles in the third week of September shows Leap Motor leading with 12,900 units sold, followed by Xiaomi and Aion [17][18] - A new autonomous driving test base was established in Jinan by Lingong Heavy Machinery and Huawei, focusing on comprehensive testing of autonomous mining vehicles [19] - ZF signed an investment agreement to expand its smart chassis localization efforts in Zhangjiagang [20] - NineSight signed a strategic agreement with Dubai's Roads and Transport Authority to launch an autonomous driving demonstration zone [21] - New Stone Technology delivered its 10,000th unmanned vehicle, marking a significant milestone in the commercialization of unmanned delivery [23] 3. Investment and Financing Events - A total of 9 financing events related to embodied intelligence were recorded from September 22 to 28, 2025, with notable investments in companies like LeXiang Technology and LuoBo Intelligent [24][34] - LeXiang Technology completed a 200 million RMB angel round led by Zhongding Capital, focusing on core component development and product scaling [24] - LuoBo Intelligent raised several million RMB in a round led by Sequoia China, targeting AI emotional companionship hardware [25] - Other companies like HeShan Technology and ShouZhi Innovation also secured funding to advance their respective technologies in humanoid robots and core components [26][27]

Content: --------- <doc id='1'>大额买入与资金流向跟踪（20250922-20250926） [Table_Authors] 郑雅斌(分析师) 本报告导读： 参考团队前期发布的专题报告选股因子系列研究（五十六、五十七），本报告旨在通 过交易明细数据构建相关指标，跟踪大额买入和净主动买入。</doc> <doc id='2'>投资要点：</doc> <doc id='3'>| | 021-23219395 | | --- | --- | | | zhengyabin@gtht.com | | 登记编号 | S0880525040105 | | | 张耿宇(分析师) | | | 021-23183109 | | | zhanggengyu@gtht.com | | 登记编号 | S0880525040078 |</doc> <doc id='4'>[Table_Report] 相关报告 量化择时和拥挤度预警周报（20250928） 2025.09.28 高频选股因子周报（20250922-20250926） 2025.09.28 低频选股因子周报（2025.09.19-2025.09.26） 2025.09.27 国内权益资产震荡，资产配置策略整体回调 2025.09.26 绝对收益产品及策略周报（250915-250919） 2025.09.24 金 融 工 程 金 融 工 程 周 报 金融工程 /[Table_Date] 2025.09.30 [Table_Summary] 个股大额买入跟踪与净主动买入跟踪。近 5 个交易日大额买入排名 前 5 的个股为：华泰股份、美凯龙、联美控股、启迪环境、酒钢宏 兴；近 5 个交易日净主动买入排名前 5 的个股为：鄂尔多斯、中信 特钢、桂冠电力、南钢股份、京泉华。（本周报中，近 5 个交易日特 指 20250922 至 20250926。） 行业大额买入与净主动买入跟踪。近 5 个交易日大额买入排名前 5 的中信一级行业为：银行、石油石化、非银行金融、钢铁、房地产； 近 5 个交易日净主动买入排名前 5 的中信一级行业为：银行、石油 石化、农林牧渔、房地产、电力及公用事业。 ETF 大额买入跟踪与 ETF 净主动买入跟踪。近 5 个交易日大额买 入排名前 5 的 ETF 为：嘉实中证 500ETF、南方标普中国 A 股大盘 红利低波 50ETF、易方达中证红利 ETF、嘉实中证稀土产业 ETF、 易方达沪深 300ETF；近 5 个交易日净主动买入排名前 5 的 ETF 为： 富国创业板 ETF、嘉实中证新能源 ETF、华夏中证内地低碳经济主 题 ETF、广发创业板 ETF、鹏华中证国防 ETF。 风险提示。因子失效风险，模型误设风险，市场系统性风险。</doc> <doc id='6'>| 1. | 跟踪指标计算说明 3 | | --- | --- | | 2. | 个股大额买入与资金流向跟踪 3 | | 3. | 宽基指数大额买入与资金流向跟踪 3 | | 4. | 行业大额买入与资金流向跟踪 4 | | 5. | ETF 大额买入与资金流向跟踪 5 | | 6. | 风险提示 5 |</doc> <doc id='7'>1. 跟踪指标计算说明 本文使用大买单成交金额占比和净主动买入金额占比，跟踪个股、宽 基指数、行业和 ETF 的大额买入与资金流向。 大买单成交金额占比刻画了大资金的买入行为。根据逐笔成交数据中 的叫买和叫卖序号，可将逐笔成交数据还原为买卖单数据，并按照每单的 成交量筛选得到大单，计算其中大买单的成交金额占当日总成交金额的比 例。 净主动买入金额占比刻画了投资者的主动买入行为。根据逐笔成交数 据中的买卖标志，可界定每笔成交属于主动买入还是主动卖出。将两者的 成交金额相减，可得净主动买入金额，并计算占当日总成交金额的比例。</doc> <doc id='8'>2. 个股大额买入与资金流向跟踪 下表为全市场大买单成交金额占比过去 5 日均值排名前 10 的股票。</doc> <doc id='9'>| | 表 1 大买单成交金额占比前 10 股票列表（20250922-20250926） | | | | | --- | --- | --- | --- | --- | | 排序 | WIND 代码 | 股票名称 | 指标值 | 时序分位数 | | 1 | 600308.SH | 华泰股份 | 89.1% | 100.0% | | 2 | 601828.SH | 美凯龙 | 88.9% | 99.2% | | 3 | 600167.SH | 联美控股 | 88.7% | 98.8% | | 4 | 000826.SZ | 启迪环境 | 88.6% | 99.6% | | 5 | 600307.SH | 酒钢宏兴 | 87.9% | 95.9% | | 6 | 601996.SH | 丰林集团 | 87.7% | 98.0% | | 7 | 601366.SH | 利群股份 | 87.7% | 99.2% | | 8 | 002453.SZ | 华软科技 | 87.7% | 99.6% | | 9 | 600935.SH | 华塑股份 | 87.5% | 97.6% | | 10 | 600606.SH | 绿地控股 | 87.3% | 99.2% | 资料来源: Wind，国泰海通证券研究 下表为部分全市场净主动买入金额占比过去 5 日均值排名前 10 的股票。</doc> <doc id='10'>| | 表 2 净主动买入金额占比前 10 股票列表（20250922-20250926） | | | | | --- | --- | --- | --- | --- | | 排序 | WIND 代码 | 股票名称 | 指标值 | 时序分位数 | | 1 | 600295.SH | 鄂尔多斯 | 17.9% | 98.4% | | 2 | 000708.SZ | 中信特钢 | 17.4% | 100.0% | | 3 | 600236.SH | 桂冠电力 | 15.2% | 98.4% | | 4 | 600282.SH | 南钢股份 | 12.7% | 95.1% | | 5 | 002885.SZ | 京泉华 | 11.6% | 100.0% | | 6 | 600323.SH | 瀚蓝环境 | 11.5% | 95.1% | | 7 | 600176.SH | 中国巨石 | 11.5% | 99.2% | | 9 | 603037.SH | 凯众股份 | 11.0% | 100.0% | | 10 | 603091.SH | 众鑫股份 | 10.9% | 99.2% | 资料来源: Wind，国泰海通证券研究</doc> <doc id='11'>3. 宽基指数大额买入与资金流向跟踪 本文使用整体法计算主要宽基指数的大买单成交金额占比和净主动买 入金额占比，列于下表。</doc> <doc id='12'>表 3 宽基指数大额买入与资金流向（20250922-20250926） | | | 大买单成交金额占比 | | 净主动买入金额占比 | | --- | --- | --- | --- | --- | | | 5 日均值 | 5 日均值分位数 | 5 日均值 | 5 日均值分位数 | | 上证指数 | 72.3% | 2.9% | -3.6% | 21.6% | | 上证 50 | 70.2% | 7.3% | 1.1% | 18.4%

Investment Rating - The report maintains an "Accumulate" rating for the company [2][6][13] Core Views - The company's Q2 2025 performance showed a decline both year-on-year and quarter-on-quarter, but the PVP application range is extensive, and the medical sector is showing signs of recovery [2][13] - The report projects a decrease in EPS for 2025 and 2026 to 0.70 and 0.82 CNY respectively, with a new EPS estimate for 2027 at 1.02 CNY [13] - The target price is set at 23.10 CNY based on a 33x PE valuation for 2025 [13] Financial Summary - Total revenue for 2023 is projected at 1,583 million CNY, with a slight increase to 1,606 million CNY in 2024, followed by a decrease to 1,520 million CNY in 2025 [4][14] - Net profit attributable to the parent company is expected to decline from 493 million CNY in 2023 to 350 million CNY in 2024, and further to 342 million CNY in 2025 [4][14] - The company’s gross margin and net margin for Q2 2025 were reported at 41.86% and 18.2% respectively, showing a decline from Q1 2025 [13] Market Data - The company's current price is 17.89 CNY, with a 52-week price range of 11.95 to 19.87 CNY [6][7] - The total market capitalization is approximately 8,695 million CNY [7] Industry Insights - PVP is widely used across various sectors including pharmaceuticals, food industry, and electronics, with applications enhancing battery performance and electronic skin sensitivity [13] - The report indicates a stabilization in PVP prices after a significant drop in H1 2025, which had previously impacted revenue and margins [13]

Group 1 - The report maintains an "overweight" rating for the automotive industry, emphasizing the potential of humanoid robots to create demand and suggesting a focus on high-cost performance technology routes similar to lithium iron phosphate cathodes in new energy vehicles [2][8] - Recommended companies include Dechang Motor Holdings, which leads in automotive motors, and Haoneng Co., which is advancing into reducers [8] - The domestic heavy truck market has shown significant growth due to the old-for-new replacement policy initiated in May, with recommendations for China National Heavy Duty Truck Group, Foton Motor, and Weichai Power [8] Group 2 - The report highlights the competitive pressure in the domestic passenger car market as the old-for-new replacement effects diminish, recommending differentiated competitors such as Great Wall Motors, SAIC Motor, and Jianghuai Automobile [8] - The report notes that the retail sales of passenger cars in China from September 1-21 reached 1.191 million units, a year-on-year increase of 1%, with cumulative retail sales for the year at 15.955 million units, up 9% [3][9] - The penetration rate of new energy vehicles in the passenger car market reached 58.5%, with retail sales of 697,000 units during the same period, reflecting a year-on-year growth of 10% [3][9] Group 3 - The humanoid robot industry is experiencing rapid iteration, with companies like Xinuo Future showing strong capabilities in core component manufacturing, including a complete production line for motors and control systems [4][10] - The report emphasizes the importance of innovation and cost advantages in the humanoid robot sector, particularly for companies like Dechang Motor Holdings [4][10]

Investment Rating - The report maintains an "Overweight" rating for the automotive industry [4][18]. Core Insights - The humanoid robot sector is gaining attention, particularly showcased at the 2025 Shanghai Industrial Expo, indicating a growing demand for innovative technologies [2][16]. - The report emphasizes the importance of identifying high-cost performance technology routes similar to lithium iron phosphate batteries in the electric vehicle sector, and companies with inherent advantages like CATL [18][17]. - The domestic passenger car market is facing increased competition as the effects of vehicle replacement programs diminish, leading to a recommendation for companies that focus on differentiated competition [18][15]. Summary by Sections 1. Industry Weekly Market Review - The automotive index remained flat over the past week, while the new energy vehicle index fell by 2%. The commercial vehicle index decreased by 3% [2][7]. - Over the past month, the automotive index increased by 5%, while the new energy vehicle index decreased by 2% [8][9]. 2. Investment Strategy and Recommendations - The report suggests focusing on companies like Dechang Motor Holdings and Haoneng Co., which are positioned to benefit from the humanoid robot supply chain [18][19]. - It highlights the potential of companies like Great Wall Motors, SAIC Motor, and JAC Motors in the competitive passenger car market [18][19]. - Recommendations for the heavy truck market include China National Heavy Duty Truck Group, Foton Motor, and Weichai Power, which are expected to benefit from industry upgrades [18][19]. - In the automotive parts sector, companies like Nexteer Automotive and Asia-Pacific Holdings are recommended due to their growth potential [18][19]. 3. Passenger Car Weekly Sales - From September 1 to 21, 2025, retail sales of passenger cars reached 1.191 million units, a year-on-year increase of 1% [15]. - The retail penetration rate for new energy vehicles reached 58.5%, with cumulative retail sales of 8.267 million units this year, reflecting a 24% year-on-year growth [15][18]. 4. Humanoid Robots at the Expo - The report notes the participation of several humanoid robot manufacturers at the Shanghai Industrial Expo, showcasing advancements in robotics technology [16][17]. - Companies like Xynova Future are highlighted for their comprehensive production capabilities in the humanoid robot sector, indicating a trend towards innovation and cost efficiency [17][18].

Fund Issuance - 12 new ESG funds were launched this month, with a total issuance of 6.21 billion units[9] - In the past year, 255 ESG public funds were issued, totaling 177.81 billion units[9] - The total number of existing ESG funds is 930, with the largest categories being ESG strategy (388 funds) and environmental protection (269 funds)[11] Market Performance - The A-share market showed signs of recovery, with the CSI 300 index rising by 0.44% and the ESG 300 index increasing by 1.07% during the week of September 22-26, 2025[5] - The average daily trading volume for the entire A-share market was approximately 2.32 trillion yuan, indicating a contraction in liquidity[5] Fund Performance - The top-performing fund last week was the Harvest Green Theme A, with a weekly return of 7.64% and a year-to-date return of 65.33%[12] - Other notable funds included the Shenwan Hongyuan New Economy A and Harvest Carbon Neutral Theme, with returns of 6.17% and 6.13% respectively for the week[12] Green Bond Issuance - A total of 141 ESG bonds were issued this month, amounting to 116.2 billion yuan[15] - Over the past year, 1,116 ESG bonds were issued, with a total value of 1,251.9 billion yuan[15] - The existing ESG bond market consists of 3,677 bonds, with green bonds making up the largest share at 2,510 bonds[15] Trading Activity - The total trading volume of ESG green bonds last week was approximately 52.78 trillion yuan, with the interbank market accounting for 75.02% of the transactions[19] - Repo transactions dominated the trading methods, comprising 94.22% of the total trading volume[22] Bank Wealth Management Products - 90 ESG bank wealth management products were launched this month, with a total of 1,087 existing products in the market[20] - The largest share of existing products is pure ESG themes, accounting for 55.47%[20] Risk Factors - Potential risks include insufficient ESG policy enforcement, lack of standardized data reporting, and lower-than-expected product issuance scales[23]

Investment Rating - The report rates the coal industry as "Overweight" [4]. Core Viewpoints - Coal prices are expected to rebound in the off-season, with pressure anticipated in the first half of 2026, but the year-on-year decline compared to 2025 will ease. It is projected that coal prices could exceed 800 RMB/ton in the second half of 2026 [2]. Summary by Sections Investment Highlights - The report recommends maintaining positions in key companies such as China Shenhua, Shaanxi Coal and Chemical Industry, and China Coal Energy, while also continuing to recommend Yanzhou Coal Mining and Jinneng Holding. The investment opportunities arising from state-owned enterprise reforms should be emphasized, which may create a sector-wide effect [4]. - The demand side shows a significant recovery, with total electricity consumption in August growing by 4.6%, compared to only 2.5% in Q1, and is expected to exceed a 5% growth rate for the year. This contradicts previous market pessimism [4]. - On the supply side, the output of raw coal in August was 390 million tons, a year-on-year decrease of 3.2%, but a month-on-month increase of 10 million tons. The total coal production for the year is expected to be stable at around 475-480 million tons, with a slight decline in H2 due to "overproduction checks" [4]. Coal Price Tracking - As of September 26, 2025, the price of Q5500 coal at Huanghua Port was 713 RMB/ton, up 0.6% from the previous week. The price of Q5000 coal at the same port was 622 RMB/ton, up 0.5% [7][10]. - The price of coking coal at Jingtang Port was 1710 RMB/ton, an increase of 6.2% from the previous week [35]. Inventory and Supply Chain - The inventory at Qinhuangdao decreased by 12.2% to 5.4 million tons as of September 25, 2025. The total inventory at northern ports was 29.64 million tons, down 0.9% [20]. - The report notes a decrease in both port and steel mill inventories, indicating a tightening supply situation [54][56]. International Coal Prices - The report highlights that Australian Newcastle coal prices have decreased, with the price of Q5500 coal at Newcastle being 71 USD/ton, up 1 USD (1.3%) from the previous week. The cost of domestic coal is lower than that of Australian imports by 7 RMB/ton [18][19].

Group 1 - The report emphasizes that the recent market adjustments present investment opportunities, and the Chinese stock market is expected to continue its upward trajectory, driven by factors such as the decline in risk-free returns and capital market reforms aimed at improving investor returns [2][3][4] - The report highlights that the Chinese economy is transitioning from a "L-shaped" recovery to a more stable growth phase, with corporate revenue and inventory growth stabilizing over the past two quarters, indicating a potential for improved asset returns and stock valuations [3][4] - The report suggests that emerging technology sectors remain a key investment focus, with recommendations for increasing allocations in cyclical financial stocks, particularly in the context of the ongoing recovery in the Hong Kong stock market [4][5] Group 2 - The transportation sector is expected to see strong performance, particularly in aviation, where demand is anticipated to surge during the upcoming holiday season, leading to optimistic profit forecasts for airlines [11][12] - The oil shipping market is experiencing a significant increase in freight rates, reaching a 30-month high, which is expected to positively impact profitability in the coming quarters [13][14] - The express delivery sector is also projected to recover profitability due to effective price increases and regulatory support against excessive competition, marking a positive outlook for Q3 [14][15] Group 3 - The report indicates that the Hong Kong stock market, particularly the Hang Seng Technology Index, is undervalued compared to historical averages, with potential for significant upward movement as technology stocks recover [28][30] - It is noted that the current price-to-earnings ratios for the Hang Seng Index and Hang Seng Technology Index are significantly lower than their peaks in 2021, suggesting room for valuation recovery [28][30] - The report anticipates that the combination of improving fundamentals and continued foreign capital inflows will support the Hong Kong market reaching new highs in the fourth quarter [31][32]

Quantitative Models and Construction Methods 1. Model Name: Weekly Rebalancing AI-Enhanced CSI 500 Wide Constraint Portfolio - **Model Construction Idea**: This model aims to enhance the CSI 500 index performance by leveraging AI-based factors while applying wide constraints on portfolio construction [72][73] - **Model Construction Process**: - The model uses deep learning factors (e.g., multi-granularity model with 10-day labels) as the basis for stock selection [72] - Constraints include: - Stock weight: 1% - Industry weight: 1% - Market cap weight: 0.3 - Turnover rate constraint: 0.3 - The optimization objective is to maximize expected returns, represented by the formula: $$ max \sum \mu_{i}w_{i} $$ where \( w_{i} \) is the weight of stock \( i \) in the portfolio, and \( \mu_{i} \) is the expected excess return of stock \( i \) [73][74] - **Model Evaluation**: The model demonstrates moderate performance under wide constraints, with cumulative excess returns shown over time [75][77] 2. Model Name: Weekly Rebalancing AI-Enhanced CSI 500 Strict Constraint Portfolio - **Model Construction Idea**: Similar to the wide constraint model but applies stricter constraints to control risk and enhance robustness [72][73] - **Model Construction Process**: - Constraints include: - Stock weight: 1% - Industry weight: 1% - Market cap weight: 0.1 - Additional constraints: - Market cap squared: 0.1 - ROE: 0.3 - SUE: 0.3 - Volatility: 0.3 - Component stock constraint: 0.8 - Optimization objective remains the same as the wide constraint model [73][74] - **Model Evaluation**: The stricter constraints result in a more stable performance, with cumulative excess returns displayed over time [76][80] 3. Model Name: Weekly Rebalancing AI-Enhanced CSI 1000 Wide Constraint Portfolio - **Model Construction Idea**: This model targets the CSI 1000 index, applying wide constraints while leveraging AI-based factors for enhanced returns [72][73] - **Model Construction Process**: - Constraints are similar to the CSI 500 wide constraint model, with a focus on smaller-cap stocks [73] - **Model Evaluation**: The model shows significant cumulative excess returns, particularly in recent years [79][86] 4. Model Name: Weekly Rebalancing AI-Enhanced CSI 1000 Strict Constraint Portfolio - **Model Construction Idea**: Similar to the wide constraint model but applies stricter constraints to manage risk and improve consistency [72][73] - **Model Construction Process**: - Constraints are similar to the CSI 500 strict constraint model, tailored for the CSI 1000 index [73] - **Model Evaluation**: The model demonstrates strong performance under strict constraints, with cumulative excess returns highlighted [85][87] --- Model Backtesting Results 1. Weekly Rebalancing AI-Enhanced CSI 500 Wide Constraint Portfolio - **Weekly Excess Return**: -1.36% (last week), -3.85% (September), 0.94% (YTD 2025) [13][78] - **Weekly Win Rate**: 23/39 weeks [13] 2. Weekly Rebalancing AI-Enhanced CSI 500 Strict Constraint Portfolio - **Weekly Excess Return**: -1.35% (last week), -1.33% (September), 3.70% (YTD 2025) [13][81] - **Weekly Win Rate**: 24/39 weeks [13] 3. Weekly Rebalancing AI-Enhanced CSI 1000 Wide Constraint Portfolio - **Weekly Excess Return**: 0.40% (last week), 0.42% (September), 9.15% (YTD 2025) [13][83] - **Weekly Win Rate**: 26/39 weeks [13] 4. Weekly Rebalancing AI-Enhanced CSI 1000 Strict Constraint Portfolio - **Weekly Excess Return**: -0.19% (last week), 0.67% (September), 14.01% (YTD 2025) [13][90] - **Weekly Win Rate**: 25/39 weeks [13] --- Quantitative Factors and Construction Methods 1. Factor Name: Intraday Skewness Factor - **Factor Construction Idea**: Captures the skewness of intraday stock returns to identify potential outperformers [6][8] - **Factor Construction Process**: Referenced in the report "Stock Selection Factor Series Research (19)" [13] - **Factor Evaluation**: Demonstrates strong performance with IC values of 0.027 (historical) and 0.042 (2025) [9][10] 2. Factor Name: Downside Volatility Proportion Factor - **Factor Construction Idea**: Measures the proportion of downside volatility in realized volatility to assess risk-adjusted returns [6][8] - **Factor Construction Process**: Referenced in the report "Stock Selection Factor Series Research (25)" [18][20] - **Factor Evaluation**: Shows moderate performance with IC values of 0.025 (historical) and 0.036 (2025) [9][10] 3. Factor Name: Post-Open Buying Intensity Factor - **Factor Construction Idea**: Quantifies the intensity of buying activity after market open to identify short-term momentum [6][8] - **Factor Construction Process**: Referenced in the report "Stock Selection Factor Series Research (64)" [22][26] - **Factor Evaluation**: Displays stable performance with IC values of 0.035 (historical) and 0.030 (2025) [9][10] 4. Factor Name: Deep Learning Factor (Improved GRU(50,2)+NN(10)) - **Factor Construction Idea**: Utilizes a gated recurrent unit (GRU) and neural network (NN) architecture to predict stock returns [6][8] - **Factor Construction Process**: Combines GRU with NN to capture temporal dependencies in high-frequency data [61][62] - **Factor Evaluation**: Strong performance with IC values of 0.066 (historical) and 0.050 (2025) [12][61] --- Factor Backtesting Results 1. Intraday Skewness Factor - **IC**: 0.027 (historical), 0.042 (2025) [9][10] - **Multi-Long-Short Return**: 3.82% (September), 16.22% (YTD 2025) [9][10] 2. Downside Volatility Proportion Factor - **IC**: 0.025 (historical), 0.036 (2025) [9][10] - **Multi-Long-Short Return**: 2.86% (September), 13.58% (YTD 2025) [9][10] 3. Post-Open Buying Intensity Factor - **IC**: 0.035 (historical), 0.030 (2025) [9][10] - **Multi-Long-Short Return**: 0.65% (September), 11.29% (YTD 2025) [9][10] 4. Deep Learning Factor (Improved GRU(50,2)+NN(10)) - **IC**: 0.066 (historical), 0.050 (2025) [12][61] - **Multi-Long-Short Return**: 2.13% (September), 7.40% (YTD 2025) [12][61]