Core Insights - The overall performance of Shenzhen state-owned enterprises' new media accounts has declined in August, with significant drops in both article output and reading volume compared to July [2][20] - The "Shenzhen State-owned Enterprises New Media Information Release Observation List" evaluates 54 accounts, including 31 primary accounts of state-owned enterprises and 23 listed companies [1][20] Group 1: New Media Performance - In August, 21 out of 31 primary accounts saw a decrease in article output, with notable reductions from Shenzhen Stock Exchange, Shenzhen Anju, and others [2][3] - Reading volume for 21 accounts also declined, with some experiencing drops of over 50%, while the Special Zone Construction Group saw a significant increase of 16,610 in reading volume [2][3] - The top three accounts by article output were Shenzhen Bus Group (71 articles), Shenzhen Energy (64 articles), and Shenzhen Metro (63 articles, with a slight increase of 1 article) [3][4] Group 2: Reading Volume and Engagement - The top three accounts by reading volume were Shenzhen Metro (361,000+), Guoxin Securities (135,000+), and Shenzhen Eastern Bus (133,000+), while some accounts had reading volumes below 1,000 [4][5] - The total number of likes was highest for Shenzhen Metro (4,079), followed by Shenzhen Bus Group (2,881) and Shenzhen Gas (1,957) [6][7] - Recommendation volume was led by Shenzhen Bus Group (1,882), Shenzhen Metro (1,252), and Shenzhen Eastern Bus (513), indicating a positive reception of their content [8][9] Group 3: Listed Companies Performance - Among the 23 listed company accounts, 14 experienced declines in reading volume and article output, indicating instability in content planning [11][20] - Shenzhen Airlines, CIMC, and China Ping An led in reading volume with 110,000+, 50,913, and 43,212 respectively, but many accounts saw significant drops [14][15] - The article output for Shenzhen Airlines was 24, while CIMC maintained a steady output, benefiting from a recent event that boosted its reading volume [14][19] Group 4: Overall Trends and Recommendations - The overall data from August indicates a disparity in performance, with top accounts maintaining strong output and engagement, while lower-tier accounts struggle [20] - Recommendations for improvement include data review, strategy adjustments, content quality enhancement, and better rhythm control to foster a more balanced development across all accounts [20]

Core Insights - The report emphasizes the need for a multi-dimensional, AI-driven framework to effectively predict and manage risks associated with short-term market surges, particularly in the context of the A-share market [2][3] - It introduces a comprehensive multi-factor system based on four core dimensions: trend, momentum, capital flow, and valuation, which collectively enhance market state characterization [2][4] - The AI-enhanced multi-factor timing strategy is expected to provide investors with an objective risk warning tool, reducing losses from blind chasing of high prices [3][4] Trend Analysis - The report illustrates that traditional indicators often fail to provide timely warnings for rapid market fluctuations driven by emotions rather than fundamentals [2][6] - The analysis of the "9·24" market surge shows that moving averages indicated a bullish trend before the surge, while subsequent signals indicated a weakening momentum [5][6][8] Momentum Indicators - The report highlights that extreme values in momentum indicators like KDJ and RSI often signal the end of a price surge, as seen during the "9·24" event where both indicators reached overbought levels [8][9] - The KDJ and RSI thresholds serve as critical points for identifying market cycles, aiding investors in timing their trades effectively [9] Capital Flow Insights - The report notes a strong correlation between trading volume and price movements during the "9·24" surge, indicating that volume often precedes price increases [11][12] - A decline in trading volume following price peaks serves as a warning signal for potential market corrections, as evidenced in the analysis [12] Valuation Metrics - The report discusses how valuation metrics, such as PE ratios, can indicate market risk accumulation, particularly when they exceed historical high thresholds [15][16] - The combination of high valuation levels and overbought momentum indicators has historically signaled market tops and subsequent corrections [15] AI-Driven Quantitative Strategy - The report outlines a comprehensive AI-driven quantitative strategy that automates the process of factor selection, modeling, and execution, enhancing the robustness of trading signals [19][20] - The strategy employs a closed-loop system that continuously optimizes itself based on real-time performance feedback, ensuring adaptability to changing market conditions [19][20] Factor Processing and Model Selection - The report emphasizes the importance of factor processing, including standardization and ranking, to ensure comparability and robustness of the indicators used in the model [30][33] - The HistGradientBoosting model is selected for its ability to capture non-linear relationships among factors, providing a more accurate timing signal for trades [39][40] Performance Evaluation - Backtesting results indicate that the AI-driven strategy significantly outperforms the market benchmark, achieving an annualized return of approximately 36.41% with a Sharpe ratio of 2.30 [41][42] - The strategy demonstrates strong risk management capabilities, maintaining a maximum drawdown of -19.51%, which is notably lower than the benchmark during volatile periods [45][46]

证券研究报告 | 2025年09月25日 AI 赋能资产配置（十七） AI 盯盘： "9·24"行情案例 策略研究·策略解读 | 证券分析师： | 王开 | 021-60933132 | wangkai8@guosen.com.cn | 执证编码：S0980521030001 | | --- | --- | --- | --- | --- | | 证券分析师： | 陈凯畅 | 021-60375429 | chenkaichang@guosen.com.cn | 执证编码：S0980523090002 | 事项： 金融市场中的短期内快速上涨行情往往因情绪驱动而非基本面改善，容易导致阶段性追高。传统技术分析 （如 KDJ、RSI、MACD、均线体系、成交量、换手率及估值水平等单一指标）虽能提供部分洞察，但其 信号纷杂、滞后性强且受主观经验影响较大，难以有效预警此类脉冲式行情的风险。 为系统性地解决这一问题，本研究旨在构建一个多维度、量化、由人工智能驱动的综合研判框架。研究首 先从趋势、动量、资金流向、估值四个核心维度出发，构建了十二个关键原始指标，形成一个全面刻画市 场状态的多因子体系，并初步判断和市场趋势的关 ...

Group 1 - The Shanghai Stock Exchange's Sci-Tech Innovation Board semiconductor materials and equipment index fell by 1.62% as of September 25, 2025, with mixed performance among constituent stocks [1] - The Sci-Tech Semiconductor ETF (588170) decreased by 1.75%, with the latest price at 1.46 yuan, while its trading volume was active with a turnover of 18.85% and a total transaction value of 267 million yuan [1] - The latest scale of the Sci-Tech Semiconductor ETF reached a new high of 1.43 billion yuan, with the number of shares also hitting a record of 967 million [1] Group 2 - According to data from SEMI, global silicon wafer shipments in Q2 2025 reached 3.327 billion square inches, marking four consecutive quarters of year-on-year growth and the highest level since Q3 2023 [2] - The semiconductor equipment and materials industry is a key area for domestic substitution, benefiting from the expansion of semiconductor demand driven by the AI revolution and advancements in lithography technology [2] - The Sci-Tech Semiconductor ETF (588170) and its linked funds focus on semiconductor equipment (59%) and materials (25%), indicating a strong emphasis on the upstream semiconductor sector [2]

Core Insights - The report from Guosen Securities highlights the increasing power of data centers and the role of liquid cooling solutions in addressing high energy consumption and heat dissipation challenges [1][2] Group 1: Industry Overview - The liquid cooling sector is in its early development stage, with various solutions having distinct advantages and disadvantages, and no established optimal solution yet [1] - According to SemiAnalysis, the global installed capacity of AI data centers is expected to reach 59 GW by 2028, potentially generating a demand for approximately 89,000 tons of cooling liquid [1] - Traditional server installations are projected to contribute an additional demand of 19,000 tons of cooling liquid [1] Group 2: Energy Consumption and Efficiency - By 2024, the total electricity consumption of computing centers in China is expected to reach 166 billion kWh, accounting for 1.68% of the total electricity consumption in society [2] - The average Power Usage Effectiveness (PUE) for ultra-large data centers in China is currently 1.46, while for large data centers, it is 1.55 [2] - New standards set by the Ministry of Finance, Ministry of Ecology and Environment, and Ministry of Industry and Information Technology require that the energy ratio of data centers not exceed 1.4 starting June 2023, and 1.3 by 2025 [2] Group 3: Liquid Cooling Technologies - The main liquid cooling technologies include cold plate and immersion cooling, with cold plate cooling being the most widely used due to its mature technology [3] - Liquid cooling offers multiple advantages such as high heat exchange efficiency, energy savings, and reduced noise [3] - Cold plate cooling is categorized into indirect and direct cooling, with indirect cooling typically being cold plate and direct cooling including immersion and spray cooling [3] Group 4: Cooling Liquid Types and Applications - Different types of cooling liquids are suited for various applications, with water-based cooling liquids used in single-phase cold plate cooling and refrigerants used in dual-phase systems [3] - Oil-based and fluorinated cooling liquids can be used in both single-phase and multi-phase immersion cooling [3] - Water-based cooling liquids are cost-effective but may promote algae and bacteria growth, while oil-based liquids have good insulation but higher viscosity [3] - Fluorinated cooling liquids, including perfluoropolyether and perfluoroamines, offer good flow properties and low toxicity but are relatively expensive [3]

Core Viewpoint - The report from Guosen Securities highlights the rapid advancement of domestic 3D models and lightweight AI models, with significant developments from Tencent, Alibaba Cloud, and Baidu, alongside a strong box office performance for the film "731" [1][3]. Group 1: Industry Performance - The media sector saw a weekly increase of 0.38%, outperforming the CSI 300 index which fell by 1.01%, but underperforming the ChiNext index which rose by 1.22% [3]. - The top gainers in the media sector included Jishi Media, Guomai Culture, and Liou Shares, while the biggest losers were Happy Blue Ocean and Shanghai Film [3]. Group 2: Key Developments - Tencent launched the mixed Yuan 3D 3.0 model, enhancing modeling precision with its 3D-DiT grading carving technology and introduced the mixed Yuan 3D Studio platform to improve production efficiency [3]. - Alibaba Cloud open-sourced the Tongyi DeepResearch model, achieving state-of-the-art results with only 30 billion parameters [3]. - Baidu's new model ERNIE-4.5-21B-A3B-Thinking topped the HuggingFace leaderboard [3]. - The film "731" grossed nearly 1 billion yuan within three days of its release [3][4]. Group 3: Investment Recommendations - The company maintains a positive outlook on the gaming sector's new product cycle and the potential recovery of the film industry, emphasizing the importance of AI applications [5]. - Recommended stocks in the gaming sector include Kaiying Network, Gigabit, and Xindong Company [5]. - In the media sector, companies like Focus Media and Bilibili are highlighted due to expected growth in advertising spending [5]. - For the film content sector, attention is drawn to Wanda Film and Huace Film for potential supply-side improvements [5]. - The report also suggests focusing on high-dividend, undervalued state-owned publishing companies [5]. Group 4: Emerging Trends - The industry continues to favor high-demand IP toys and AI applications in their early stages [2]. - Recommended stocks in the IP toy sector include Pop Mart, Zhejiang Digital Culture, and Yaoji Technology [2]. - Key application areas for AI include gaming, toys, advertising, education, e-commerce, and social media, with specific focus on 2B/2G and 2C opportunities [2].

Core Viewpoint - The article emphasizes the importance of digital transformation in the securities industry, highlighting it as a critical factor for survival and development, driven by the need to enhance service efficiency, risk management, and business innovation [2]. Group 1: Digital Transformation Insights - The central theme of the "2025 China Securities Industry Digital Practice Summit" is "Intelligent Wealth Together, Win the Future," showcasing the industry's focus on digital finance [2]. - The Central Financial Work Conference has recognized "digital finance" as a key component in building a strong financial nation, indicating the strategic importance of digital transformation in the securities sector [2]. - AI is viewed as a transformative force, not just for efficiency but for creating new possibilities, necessitating a comprehensive shift in organizational structure and operational models [2][3]. Group 2: Awards and Recognitions - The "2025 China Securities Industry Digital Practice Jun Ding Award" was announced, recognizing outstanding institutions in digital transformation [4]. - The winners of the "2025 China Securities Industry Digital Practice All-round Jun Ding Award" include major firms such as Huatai Securities, Guotai Junan Securities, and CITIC Securities, among others [6]. - The "2025 China Securities Industry Digital Practice Newcomer Jun Ding Award" recognized firms like Zhongyou Securities and Huafu Securities, highlighting emerging players in the digital landscape [7]. Group 3: Case Studies and Innovations - The article mentions various innovative digital practices and platforms developed by securities firms, such as AI-driven investment models and intelligent advisory products [11][12]. - Specific case studies include the "AI Wenchang Star" investment banking model application platform and a low-latency options market-making platform, showcasing the technological advancements in the industry [10][11].

Group 1: Market Overview - The report highlights a decline in major market indices, with the Shanghai Composite Index closing at 3821.83 points, down by 0.17% [2] - The trading volume across various indices was significant, with a total transaction amount of 10716.97 billion yuan for the Shanghai Composite Index [2] Group 2: Wealth Management and Asset Allocation - The report discusses the trend of "deposit migration," where residents are reallocating their wealth from low-yield deposits to riskier assets, significantly impacting asset prices [5][6] - It notes that as of June 2025, deposit rates fell below 2.0%, prompting a shift in wealth allocation towards risk assets, which has been a driving force behind the current capital market rally [6] - The analysis identifies three key factors influencing changes in resident wealth: aging population, increased wealth concentration among high-net-worth individuals, and a shift towards asset management for wealth allocation [6] Group 3: Asset Management Products - The report emphasizes the expansion of asset management products in response to declining risk-free rates and the need for diversified investment solutions [7] - It highlights the importance of tailoring products to meet the varying risk appetites of investors, with a focus on stable and controlled return products for lower-risk tolerance investors [7] Group 4: Robotics and AI Infrastructure - The report notes significant investments in the robotics sector, particularly Figure's completion of over 1 billion USD in funding, indicating strong growth potential in humanoid robotics [9] - It also mentions Microsoft's commitment to invest 300 billion USD in AI infrastructure in the UK, reflecting the ongoing demand for AI-driven cloud solutions [10] Group 5: Real Estate Market Analysis - The report presents a stark contrast between the stock market, which has risen by 53% since September 2024, and the real estate market, which has seen a decline of 4.4% in housing prices [15][16] - It discusses the lack of new catalysts for the real estate market post-March 2025, leading to a divergence in performance between the stock and real estate markets [15] - The report suggests that short-term improvements in housing prices may rely on new narratives or policies rather than existing measures [16]

Group 1 - The core viewpoint of the report is that Tongguan Gold (00340) is rated "outperform" by Guosen Securities, with projected net profits of HKD 780 million, HKD 1.05 billion, and HKD 1.26 billion for 2025-2027, representing growth rates of 269%, 34%, and 21% respectively [1] - The company aims to enhance research on mineralization patterns in Gansu and Tongguan, actively pursue external acquisitions for growth, and establish strategic cooperation with Zijin Mining [1][2] - The company is characterized as a small but beautiful regional gold mining enterprise, with a focus on gold mining and recovery, holding an average resource grade of 8.26 grams per ton and a resource volume of 55 tons as of the end of 2024 [2] Group 2 - Gold prices are expected to remain strong due to weakening US dollar credit, with the dollar index having dropped by 10% year-to-date as of mid-September [2] - Central bank gold purchases have exceeded 1,000 tons for three consecutive years, significantly higher than the average of 473 tons from 2010 to 2021, providing long-term support for gold prices [2] - The company has signed a long-term gold streaming agreement with Zijin Mining, receiving USD 25 million in upfront funding while committing to deliver approximately 422 kilograms of gold over nine years [4] Group 3 - The company is focused on resource expansion, with both production and reserve mines being prioritized [3] - The Tongguan mining area currently has only Xiangshun Mining in production, with a processing plant capable of handling 1,500 tons, while the Gansu area is supported by Beidong Mining, which is gradually reaching production capacity [3] - The company aims to achieve a total processing capacity of 940,000 tons annually across both mining areas, with significant growth potential remaining [3]

Group 1: AIDC Power Equipment and Grid Industry - The AIDC power equipment sector has seen a general increase in the past two weeks, with the top three performers being uninterruptible power supplies (UPS) at +21.0%, high voltage direct current (HVDC) at +20.7%, and battery backup power (BBU) at +15.9% [1] - Recommendations for investment focus include four key areas: transformers and switchgear, UPS and HVDC, active power filters (APF), and server power supplies, with specific companies suggested such as Jinpan Technology, Mingyang Electric, Hezhong Electric, Shenghong Co., and Weilan Lithium [1] - For the grid sector, the national power engineering investment completion amount in July 2025 was 65.3 billion yuan, down 8.9% year-on-year, while the cumulative investment from January to July was 428.8 billion yuan, up 3.1% year-on-year [1] Group 2: Wind Power Industry - In July 2025, China's newly installed wind power capacity was 2.28 GW, a decrease of 44.0% year-on-year, while the cumulative new capacity from January to July reached 53.67 GW, an increase of 79.4% year-on-year, totaling 574.87 GW, which accounts for 15.7% of total installed capacity [1] - The wind power sector has also experienced a general increase in the past two weeks, with the top three performing segments being bearings at +14.9%, complete machines at +12.5%, and blades at +11.7% [1] Group 3: Offshore Wind and Onshore Wind Developments - Major projects in Jiangsu and Guangdong for offshore wind are set to commence in the first half of 2025, with expectations for planning, competitive allocation, bidding, and policies to be implemented in the second half of the year [2] - The annual average offshore wind installation during the 14th Five-Year Plan period is expected to exceed 20 GW, significantly surpassing the previous plan's levels [2] - The onshore wind industry is projected to reach 100 GW of installed capacity in 2025, marking a historical high, with component manufacturers experiencing simultaneous increases in volume and price, leading to substantial growth in annual performance [2] Group 4: Public Utilities and Environmental Protection - In August, the industrial power generation volume increased by 1.6% year-on-year, totaling 936.3 billion kWh, while the cumulative generation from January to August was 6419.3 billion kWh, reflecting a 1.5% year-on-year growth [3] - The Guangdong Provincial Development and Reform Commission and Energy Bureau released a plan to deepen the market-oriented reform of renewable energy grid pricing, with existing project pricing set at 0.453 yuan per kWh [3]