Core Insights - The probability of interest rate cuts within the year has decreased, with demand-side policies potentially sustaining inflation expectations, leading to continued pressure on the bond market and further postponement of the buying window [1][3][28] - The personal consumption loan interest subsidy policy is expected to compress the central bank's total monetary policy easing space, with the next potential rate cut window likely pushed to the first quarter of 2026 [3][12][28] - Inflation expectations have been reinforced under the "anti-involution" policy, with demand-side policies providing better support for price increase expectations, leading to a resurgence in "anti-involution trading" [2][20][28] Summary by Sections Understanding Personal Consumption Loan Interest Subsidy Policy - The policy covers personal consumption loans under 50,000 yuan and key areas such as automotive, healthcare, and education, with a maximum subsidy of 3,000 yuan for eligible borrowers [10][11] - The subsidy is equivalent to a targeted interest rate cut of 0.5-1 percentage points, reducing the effective financing cost for consumers [11][12] Impact on Bond Market - The bond market's buying window is likely to be further delayed due to the subsidy policy, which compresses the central bank's monetary policy easing space [3][28] - The anticipated inflation expectations and strong performance in the equity market may continue to exert pressure on the bond market [28] Inflation Expectations and Market Dynamics - The "anti-involution" policy has led to a self-reinforcing cycle of price signals and inflation expectations, with significant price increases in sectors like coal and steel [2][20] - The market has begun to bet on a re-inflation scenario, with indicators showing a shift in investor sentiment towards inflationary expectations [20][27]

证券研究报告 | 浙商早知道 报告日期：2025 年 08 月 13 日 浙商早知道 2025 年 08 月 13 日 ： 张延兵 执业证书编号：S1230511010020 ： 021-80106048 ： zhangyanbing@stocke.com.cn 市场总览 重要点评 ❑ 【浙商机械 邱世梁/王华君/何家恺】中力股份（603194）公司点评：与白犀牛战略合作共建智慧物流体系，智能 搬运机器人打开空间——20250812 http://www.stocke.com.cn 1/3 请务必阅读正文之后的免责条款部分 1、主要事件 与白犀牛战略合作共建智慧物流体系，智能搬运机器人打开空间 ❑ 大势：周二上证指数上涨 0.5%，沪深 300 上涨 0.5%，科创 50 上涨 1.9%，中证 1000 上涨 0.3%，创业板指上涨 1.2%，恒生指数上涨 0.3%。 ❑ 行业：周二表现最好的行业分别是通信（+2.2%)、电子（+1.9%)、煤炭（+1.0%)、家用电器（+0.7%)、房地产 （+0.7%)，表现最差的行业分别是国防军工（-1.0%)、钢铁（-0.8%)、建筑材料（-0.5%)、食品饮料（-0 ...

证券研究报告 | 深度研究 | ESG 及绿色金融 以责促优，共筑价值：公募基金 ESG 尽责管理新范式（一） 核心观点 尽责管理本质上是投资者结构变迁下的公司治理新模式。国内公募基金管理人在短中 期内应采取软性治理策略，并尽可能地成为上市公司治理中"具有互惠性质的积极参 与者"，降低公司治理环节的内部摩擦。公募基金尽责管理现阶段的主要任务和最具普 适性的议题应集中在提升公司治理水平，特别是中小股东权利保护上。 ❑ 尽责管理本质上是投资者结构变迁下的公司治理新模式 以美国市场为代表的资管机构主导公司治理模式的形成，是资本市场长期发展过 程中满足较多严苛条件下的产物。但可能也是现阶段可实现"覆盖度广+影响力高" 的一种上市公司治理模式，对我国资本市场和公募基金高质量发展有重要的参考 意义。 ❑ 公募基金尽责管理三大问：委托代理、投资者结构和组合分散化 委托代理问题将因监管要求和尽责管理的投资价值而得到有效缓解。目前 A 股公 司治理模式的最大特点是混合了"控股大股东+专业机构+居民部门"，这一模式并 非终局，而仍在动态演变中。被动指数化会导致低换手率和退出减少，公募基金 逐渐成为耐心资本，实现"组合分散+股东权 ...

Investment Rating - The investment rating for the company is "Buy" (maintained) [5] Core Views - The company achieved a revenue of 3.495 billion yuan in the first half of 2025, representing a year-on-year growth of 21.38%, while the net profit attributable to shareholders was a loss of 170 million yuan, reducing losses by 63.82% [1] - The company has seen significant growth in its four main business areas: automotive, industrial control, consumer electronics, and AI-related applications, contributing to a steady increase in revenue [2] - The transition to a system-level foundry model has enhanced the company's growth potential, allowing it to provide comprehensive chip system services, which has led to a substantial increase in revenue from module packaging [3] - Continuous breakthroughs in product development and market expansion, particularly in the AI sector, have positioned the company favorably for future growth [4] Summary by Sections Financial Performance - In Q2 2025, the company reported a revenue of 1.762 billion yuan, a year-on-year increase of 15.39%, and achieved a net profit of 12 million yuan, marking a return to profitability [1] - The company expects revenues for 2025-2027 to be 8.411 billion yuan, 10.605 billion yuan, and 13.256 billion yuan, respectively, with corresponding net profits of -446.71 million yuan, 662.39 million yuan, and 1.33282 billion yuan [8] Business Segments - Revenue from automotive, industrial control, and consumer electronics grew by 23%, 35%, and 2% respectively in the first half of 2025, with AI-related applications contributing 196 million yuan, accounting for 6% of total revenue [2] - The module packaging business saw a revenue increase of 141%, with automotive power module revenue growing over 200% [3] Market Position and Strategy - The company has successfully transitioned to a system-level foundry model, enhancing its service offerings and market recognition, particularly in automotive and AI sectors [3] - The company is actively expanding its presence in emerging markets, with significant advancements in automotive and data center technologies [4]

Market Overview - On August 11, the Shanghai Composite Index rose by 0.34%, the CSI 300 increased by 0.43%, the STAR 50 climbed by 0.59%, the CSI 1000 went up by 1.55%, the ChiNext Index surged by 1.96%, and the Hang Seng Index gained 0.19% [3][4] - The best-performing sectors on August 11 were power equipment (+2.04%), communications (+1.95%), computers (+1.94%), electronics (+1.76%), and food and beverage (+1.45%). The worst-performing sectors included banking (-1.01%), oil and petrochemicals (-0.41%), coal (-0.35%), utilities (-0.31%), and transportation (-0.19%) [3][4] - The total trading volume for the A-share market on August 11 was 1,849.9 billion yuan, with net inflow from southbound funds amounting to 38.34 million Hong Kong dollars [3][4] Key Insights - The report indicates that the A-share market is currently experiencing its first "systematic slow bull" market in history. The long-term target for the Shanghai Composite Index is likely no longer limited to challenging the 3,674 high point, with a recommendation to focus on "large finance + pan-technology" sectors [5][6] - Since the initiation of the stock split reform in April 2005, the A-share market has undergone four bull markets, with the first three being "systematic bull markets" characterized by steep upward slopes. The fourth was a "structural bull market" with a more gradual increase, driven primarily by capital market reforms and liquidity easing [5][6] - The report suggests that the market began a long-term bottoming process in 2024, with a strong market performance following April 7, 2025, marking the start of the fifth bull market in A-share history. The current "systematic slow bull" is driven by improved risk appetite and declining risk-free interest rates, alongside China's rise and advantages, creating a "slow" bull market structure [5][6] Sector Focus - The report emphasizes a focus on sectors with external advantages and improving prosperity, such as innovative pharmaceuticals and new energy, as well as defensive sectors that serve as "ballast," particularly banks [6]

WAIC 2025 上，AI 在多场景快速落地，展现强劲算力需求与商业潜力。京东发 布 JoyAI 大模型，参数覆盖 3B 至 750B，已应用于物流、零售、供应链等垂直领 域，显著提升推理效率与行业赋能。具身智能成为热点，商汤发布"悟能"平台， 融合大模型与世界模型，推动机器人具备真实环境中的感知、理解与执行能力。 自动驾驶加速落地，WeRide 获 L4 Robotaxi 运营牌照，斑马智行展示多模态智能 座舱，萝卜快跑接入 Uber 全球网络，部署千辆无人车。高并发、低延迟、端侧 部署等新需求为公司提供广阔市场空间与落地机遇。 ❑ 中美贸易战带来结构性利好，出口管制倒逼本土创新崛起 美国持续收紧高性能 AI芯片出口，限制 A100、H800等型号，制约国内训练与推 理能力。但这也为国产芯片提供替代窗口。公司具备自研指令集与微架构能力， 能绕开关键依赖，成为"去美化"背景下的重要支撑。尽管 H20 放开出口将短期 加剧竞争，但中长期看，自主可控仍是主线，政策与应用倾斜将持续利好本土厂 商。行业出清有望加快，推动缺乏核心技术的企业出局，有助于公司进一步巩固 市场地位。在政策支持、需求升级与客户信任增强下，公 ...

Investment Rating - The report assigns a "Buy" rating for the company [5][72]. Core Views - The company's gaming business, led by its wholly-owned subsidiary Kemu Games, has a strong track record in overseas game publishing, with past titles like "Age of Origins" and "War and Order" generating revenues exceeding 30 billion and 10 billion respectively. Future titles such as "Stellar Sanctuary" and "Next Agers" are expected to exceed revenue expectations [1][4]. - The company is positioned in the low-performing segment of the gaming sector for 2025, with its current stock price reflecting only the expected performance of existing games and not accounting for new product lines [3][4]. - The company is projected to achieve revenues of 69.3 billion, 89.4 billion, and 99.3 billion for 2025, 2026, and 2027 respectively, with net profits of 12.9 billion, 19.6 billion, and 21.5 billion [5][72]. Summary by Sections Company Overview - The company, established in 1998 and listed in 2009, aims to be a leader in the digital economy with a focus on game development and AI technology [18][19]. - Its gaming business is a top-tier player in the domestic mobile game market, with over 250 million registered users across more than 150 countries [19]. Industry Analysis - SLG games are characterized by high entry barriers and sustained revenue generation, with a typical lifecycle of 6 to 8 years [26]. - The "SLG+X" model, which integrates various gameplay elements, has gained popularity and is expected to enhance user acquisition and retention [35]. Financial Performance - The company has shown significant revenue growth, with a peak revenue of 19.4 billion in Q4 2024 and a record net profit of 4.63 billion in Q3 2024 [23][25]. - The overall expense ratio has improved, with a notable decrease in sales expense ratio from 54.5% in Q1 2021 to 43.5% in Q1 2025 [25]. New Product Development - The upcoming titles "Stellar Sanctuary" and "Next Agers" are set to initiate a new product cycle, with "Stellar Sanctuary" expected to contribute significantly to revenue [53][66]. - The company has multiple SLG games in development, with plans for commercial testing in the near future [60]. Valuation and Market Potential - The estimated target market capitalization for 2026 is around 400 billion, based on a P/E ratio of 20 times projected earnings [4][72]. - The global SLG game market is anticipated to exceed 20 billion USD by 2027, indicating strong growth potential for the company [30].

Quantitative Models and Construction Methods 1. Model Name: Multivariate CNN-LSTM - **Model Construction Idea**: The model leverages the advantages of CNN and LSTM in different scenarios to predict multiple parallel financial time series by considering the correlation between them[12][14]. - **Detailed Construction Process**: - **General Structure**: The model consists of an input layer, a one-dimensional convolutional layer, a pooling layer, an LSTM hidden layer, and a fully connected layer to produce the final prediction results[14]. - **Formula**: $$ {\hat{x}}_{k,t+h}=f_{k}(x_{1,t},\dots,x_{k,t},\dots,x_{1,t-1},\dots,x_{k,t-1},\dots) $$ This formula indicates that each variable depends not only on its past values but also on the past values of other variables[11]. - **Hyperparameters**: - Number of convolution filters: 64 - Convolution kernel size: 2 - Use of padding: Yes - Pooling layer window size: (2,2) - Number of hidden units in the first LSTM layer: 128 - Number of hidden units in the second LSTM layer: 128 - Activation method between LSTM layers: ReLU - Time series look-back window: 10 - Number of training epochs: 100[20] - **Evaluation Metric**: Root Mean Square Error (RMSE) $$ RMSE={\sqrt{\frac{1}{n}\sum_{i}({\hat{y_{i}}}-y_{i}\,)^{2}}} $$ where \( y_i \) represents the standardized index price, and \( \hat{y_i} \) represents the CNN-LSTM prediction value[21]. - **Model Evaluation**: The model achieved good tracking and high accuracy in predicting multiple parallel financial time series, similar to the performance in predicting stock indices in the Asia-Pacific market[14][17]. 2. Model Name: Grouped Multivariate CNN-LSTM - **Model Construction Idea**: To improve prediction accuracy, the industry indices are grouped based on investment attributes, and a separate prediction model is constructed for each group[26][27]. - **Detailed Construction Process**: - **Grouping**: The industry indices are divided into six groups: Consumer and Medicine, Upstream Resources and Materials, High-end Manufacturing, Real Estate and Infrastructure, Big Tech, and Big Finance[27]. - **Model Structure**: Each group of industry indices is predicted using a separate CNN-LSTM model, as shown in the general structure diagram[28]. - **Evaluation Metric**: The prediction accuracy is evaluated using RMSE, similar to the original model[33]. - **Model Evaluation**: Grouping and training different CNN-LSTM sub-models for each industry group improved the prediction accuracy, especially for industries with previously low prediction accuracy[30][32]. Model Backtesting Results 1. Multivariate CNN-LSTM Model - **Prediction Error (Training Phase)**: 1.52% to 3.18%[23] - **Prediction Error (Testing Phase)**: 1.56% to 3.30%[23][25] 2. Grouped Multivariate CNN-LSTM Model - **Prediction Error (Training Phase)**: 1.49% to 2.60%[33] - **Prediction Error (Testing Phase)**: 1.61% to 2.82%[33] Quantitative Factors and Construction Methods 1. Factor Name: Weekly Industry Rotation Signal - **Factor Construction Idea**: Use the predicted values from the multivariate CNN-LSTM model to estimate the future weekly returns of industry indices and select the top five industries with the highest expected returns for equal-weight allocation[3]. - **Detailed Construction Process**: - **Prediction**: Predict the future weekly returns of industry indices using the multivariate CNN-LSTM model[34]. - **Allocation**: Every five trading days, select the top five industries with the highest expected returns for equal-weight allocation[35]. - **Training**: Retrain the model at the beginning of each quarter using an extended window of historical data from March 2014 to the training point[35]. - **Factor Evaluation**: The annualized return of the industry rotation portfolio reached 15.6%, with an annualized excess return of approximately 11.6%, and the risk-return characteristics significantly improved compared to the benchmark[3][35]. Factor Backtesting Results 1. Weekly Industry Rotation Signal - **Annualized Return**: 15.6%[38] - **Annualized Volatility**: 25.6%[38] - **Maximum Drawdown**: -27.1%[38] - **Sharpe Ratio**: 0.7[38] - **Longest Drawdown Recovery Time**: 248 days[38]

Investment Rating - The industry investment rating is optimistic [1] Core Insights - The financial IT sector shows significant elasticity during the initial stages of a bull market, with notable price increases and valuation expansions [3] - The combination of technology and finance attributes leads to a "Davis Double Play" effect during bull markets, particularly highlighted in 2015 [4] - Current advancements in AI and new business developments are expected to drive further growth in the financial IT sector [5] Summary by Sections 2014-2015: Liquidity Explosion, Financial Technology Leads - The bull market from 2014 to 2015 was driven by ample liquidity and the rise of mobile internet, leading to significant gains in financial technology stocks [15][19] - Financial technology stocks experienced substantial price increases, with some stocks seeing gains close to 450% compared to mid-2014 levels [4] - The financial IT sector benefited from increased investor participation and software usage during the bull market [33] 2016-2018: Structural Bull Market, Varied Performance in Financial Technology - The period from 2016 to 2018 was characterized by a structural bull market influenced by supply-side reforms and foreign capital inflows [43] - Financial technology stocks underperformed compared to the broader market during this period, primarily due to high valuations and changing market preferences [46][52] - The financial IT sector faced challenges as the market shifted focus towards blue-chip and consumer stocks, leading to a decline in growth stocks [56] 2019-2021: Core Assets Drive Structural Bull Market - The financial technology sector saw a resurgence from 2019 to 2021, driven by global liquidity and domestic industrial upgrades [70] - The introduction of the Sci-Tech Innovation Board in 2019 significantly boosted the financial technology sector, with strong performance noted in various market phases [76][81] - Financial technology stocks outperformed the market during key periods, reflecting the sector's recovery and growth potential [82]

Investment Rating - The investment rating for the company is maintained as "Buy" [7] Core Views - The company achieved a revenue of 3.912 billion yuan in the first half of 2025, representing a year-on-year increase of 13.70%. The net loss attributable to the parent company was 226 million yuan, with a net loss of 232 million yuan after deducting non-recurring items. In Q2 2025, the company recorded a revenue of 2.369 billion yuan, a year-on-year increase of 38.20%, with a net loss of 28 million yuan, significantly reduced from approximately 216 million yuan in Q1 [1][2] - The rapid improvement in the company's Q2 performance reflects the industry's supply-demand dynamics, with expectations for continued revenue and profit growth in the latter half of the year due to favorable industry trends [2][3] - The global storage market is expected to recover, with a market size reaching 165.52 billion USD in 2024, a year-on-year increase of 79.3%. The demand for storage driven by AI technology is anticipated to provide significant growth opportunities for the semiconductor storage industry [3][4] Financial Summary - Revenue forecasts for 2025-2027 are 8.5 billion yuan, 10.8 billion yuan, and 14.2 billion yuan, respectively, with corresponding net profits of 420 million yuan, 1.13 billion yuan, and 1.58 billion yuan. The company maintains a "Buy" rating based on these projections [5][12]