Investment Rating - The report initiates coverage with an "Accumulate" investment rating for the company [1][8]. Core Views - The company is a leading player in Anhui's thermal power sector, with a strong supply-demand balance supporting high utilization hours. Expansion into Xinjiang is expected to significantly enhance profits. The company has diversified its energy sources, including coal, wind, pumped storage, and nuclear power, making investment returns a crucial pillar for its performance [1][7][8]. Summary by Sections Section 1: Anhui Thermal Power Leader - The company is the leading thermal power operator in Anhui, with a total installed capacity of 13.66 million kW as of the end of 2024, primarily from coal power [7][12]. - Revenue has grown from 16.09 billion yuan in 2019 to 30.09 billion yuan in 2024, with a CAGR of 13.34% [7][12]. - The company’s net profit for 2023 and 2024 is projected to be 1.43 billion yuan and 2.06 billion yuan, respectively, reflecting significant year-on-year growth [7][8]. Section 2: Strategic Expansion in Xinjiang - The company has strategically expanded into Xinjiang, where its power plants benefit from low coal costs due to proximity to coal fields. The net profit from these plants is expected to contribute significantly to overall performance [7][61][62]. - The company’s Xinjiang plants are projected to generate a net profit of 594 million yuan in 2025, accounting for 28.8% of the company's total profit [7][8]. Section 3: Diverse Energy Investments - The company has diversified its energy portfolio, with 7.98 million kW of equity capacity in various energy sources, including coal, wind, and nuclear power [7][8]. - Investment returns from these diverse sources are becoming increasingly important for the company's overall performance stability [7][8]. Section 4: Profit Forecast and Valuation - The company’s projected net profits for 2025, 2026, and 2027 are 2.29 billion yuan, 2.44 billion yuan, and 2.53 billion yuan, respectively, with corresponding PE ratios of 8.0x, 7.5x, and 7.3x [8][12].

Investment Rating - The report maintains a "Strong Buy" rating for several key companies in the military industry, including 中航西飞, 中航光电, 航天彩虹, and others, indicating a strong expectation for their stock price to outperform the benchmark index by over 20% [8]. Core Insights - The military industry is highlighted as a crucial area for investment, particularly in aerospace and defense sectors, with a focus on companies that are positioned to benefit from increased military spending and technological advancements [2][3]. - The report emphasizes the potential of commercial aerospace, particularly low-orbit satellite applications, which are expected to have significant military and civilian applications [3]. - The upcoming Dubai Airshow is anticipated to be a major event, showcasing over 200 aircraft and new technologies, which could serve as a catalyst for the industry [20][22]. Industry Overview - The military industry index has shown a performance of -3.8% over the past month, but a positive trend of 12.1% over six months and 13.2% over twelve months, indicating a recovery and growth potential [5][11]. - As of November 14, 2025, the SW National Defense and Military Industry Index has increased by 13.35% year-to-date, although it has underperformed the Shanghai and Shenzhen 300 Index by 4.27% [11][12]. - The report notes that the industry consists of 119 listed companies with a total market capitalization of approximately 2060.9 billion [3]. Key Events and Catalysts - Significant geopolitical events, such as Japan's escalating rhetoric regarding Taiwan, are noted as influencing factors in the military sector, potentially impacting defense spending and strategic priorities [18][19]. - The successful completion of the first sea trial of the Navy's 四川舰 (Sichuan Ship) is highlighted as a key development, showcasing advancements in naval capabilities [22]. - The report identifies several key companies that are expected to benefit from increased demand in military trade and domestic military needs, including 中航沈飞, 中航西飞, and others [7].

Investment Rating - The report maintains a "Recommended" rating for the coal mining industry, indicating a positive outlook for the sector based on fundamental conditions [5]. Core Views - The report highlights that while there has been a slight price correction in thermal coal, the fundamental supply-demand dynamics remain unchanged. Supply is slightly contracting due to safety inspections and maintenance in coal mines, while demand from non-electric sectors like metallurgy and chemicals remains stable. However, some traders are slowing their purchasing pace due to stable prices at northern ports [11][12]. - The report notes a year-on-year decline of 2.3% in China's raw coal production for October, with a significant increase in thermal power generation by 7.3% year-on-year. The upcoming cold weather is expected to increase coal consumption at power plants, potentially leading to a rise in market coal prices [11][16]. - The report also mentions that the coking coal market is experiencing weaker prices, with a shift towards on-demand purchasing as high-priced transactions show signs of fatigue. The overall sentiment in the coking coal market is subdued, with prices expected to remain stable in the short term [11]. Summary by Sections 1. Investment Views - The report indicates that the supply side is tightening, with a continued expectation of supply contraction. The demand side is also showing resilience, particularly with the anticipated increase in coal consumption due to colder weather [11][12]. 2. Coal Sector Performance and Stock Review - The coal mining index has shown a decline of 1.54%, with major coal companies experiencing mixed performance. Notable gainers include Antai Group (+57.29%) and Dayou Energy (+22.32%), while Huayang Co. (-6.17%) and Lu'an Environmental Energy (-5.39%) faced significant declines [12][13]. 3. Important Announcements and Industry News - National coal production data for October shows a total output of 40.675 million tons, marking a 2.3% year-on-year decrease. This is the fourth consecutive month of decline [16]. - A significant acquisition was finalized between Hengyuan Coal Power and Shaanxi Black Cat, involving the transfer of 100% equity in two coal companies for a total of 440 million yuan [17]. 4. Dynamic Data Tracking - As of November 14, the average price for thermal coal at Qinhuangdao Port was reported at 832.0 yuan/ton, reflecting a week-on-week increase. Meanwhile, the inventory levels at major ports are showing slight increases, indicating a mixed market sentiment [3][4][19]. 5. Key Company Valuations - The report provides detailed valuations for key companies in the coal sector, highlighting their market capitalizations and projected earnings. For instance, China Shenhua has a market cap of 822.8 billion yuan with a projected PE ratio of 14.0 for 2024 [43].

Investment Rating - The report maintains a positive investment rating for the restaurant supply chain sector, suggesting increased attention due to signs of demand recovery and growth potential for quality companies [2][38]. Core Insights - The restaurant supply chain sector is experiencing a shift where companies are transitioning from being mere supporters to active drivers of innovation and demand, highlighting the importance of R&D and innovation capabilities [10][22]. - The industry is witnessing a structural opportunity as the chain restaurant rate continues to rise, with expectations for further growth in the coming years [18][10]. - Current valuations in the sector are at historically low levels, indicating potential for recovery as demand improves [30][34]. - The report emphasizes the importance of mergers and acquisitions as companies seek to enhance their competitive positions and bind key customers [22][24]. Summary by Sections Industry Status - Overall demand in the restaurant sector remains weak, but signs of recovery are evident, particularly during holiday periods [10][11]. - The restaurant supply chain industry is projected to maintain a compound annual growth rate of over 15% in the next three years, outperforming the broader restaurant market [14][10]. Company Changes - Companies are increasingly focusing on R&D and innovation to meet the evolving demands of chain restaurants, which require standardized and stable supply [22][10]. - Mergers and acquisitions are being utilized to strengthen customer relationships and enhance resource capabilities [24][22]. - New retail channels are being explored to drive growth, with companies expanding into high-end and online markets [25][10]. Valuation Analysis - The current valuation of the sector is below the 20th percentile of the past decade, suggesting significant upside potential as demand recovers [34][30]. - The report notes that the valuation decline over the past five years has been primarily due to reduced demand and high initial valuations [30][34]. Investment Recommendations - The report suggests increasing focus on specific companies such as Haidilao, Angel Yeast, and others, which are expected to benefit from demand recovery and improved operational performance [38][39].

Investment Rating - The report maintains a "Strong Buy" investment rating for Tencent Holdings [2][6] Core Insights - Tencent's Q3 2025 performance exceeded expectations with revenue of 192.9 billion yuan, a year-on-year increase of 15% and a quarter-on-quarter increase of 5% [1] - The company's AI ecosystem continues to accelerate, enhancing various business lines including gaming, advertising, and social networking [6][39] - The report forecasts revenue growth for 2025-2027, projecting revenues of 753.5 billion, 832.5 billion, and 907.7 billion yuan respectively, with adjusted net profits of 260.2 billion, 293.6 billion, and 325.4 billion yuan [6][39] Business Segment Performance 1. Online Gaming - Q3 2025 revenue from online gaming reached 63.6 billion yuan, a 23% year-on-year increase, driven by strong domestic and international performance [10][14] - Domestic gaming revenue grew 15% to 42.8 billion yuan, while international gaming revenue surged 43% to 20.8 billion yuan [10][16] 2. Social Networking - Social networking revenue for Q3 2025 was 32.3 billion yuan, a 5% year-on-year increase, slightly below expectations [18] - The growth was primarily driven by paid music subscriptions and video live streaming services [18] 3. Marketing Services - Marketing services revenue reached 36.2 billion yuan in Q3 2025, a 21% year-on-year increase, benefiting from improved ad exposure and AI-driven optimization [23][24] 4. Financial Technology and Enterprise Services - Revenue from financial technology and enterprise services was 58.2 billion yuan, a 10% year-on-year increase, supported by growth in commercial payments and demand for AI-related services [29] 5. Profitability and AI Integration - The company achieved a gross profit of 108.8 billion yuan in Q3 2025, with a gross margin of 56.4%, reflecting improvements in high-margin businesses [36] - Significant investments in AI are expected to enhance operational efficiency and drive future growth across various segments [36][39]

========= Content: --------- <doc id='2'>任瞳 S1090519080004 rentong@cmschina.com.cn 刘凯 S1090524120001 liukai11@ cmschina.com.cn 杨航 S1090523010004 yanghang4@cmschina.com.cn 证券研究报告 | 金融工程 2025 年 11 月 15 日 专题报告 ❑ 成长因子表现：以招商证券量化团队因子库中的成长因子为例，对其有效 性进行测试。综合来看，标准化预期外盈利因子（SUE）表现较为优异。 除此以外，单季度 ROE 同比因子表现同样较优。相关性方面，净利润同 比加速度因子、标准化预期外收入因子与其他因子的相关性相对较低。 ❑ 假设已知下一期上市公司的净利润增速，从而构建净利润增长的投资组 合，先验测算该组合的业绩表现。已知未来成长确定性的未来成长组合能 够稳定的战胜历史增长组合，2010 年以来每年均能战胜历史成长组合。 ❑ 通过上市公司业绩增长的转移矩阵可以发现，当前的业绩高增速仅能部分 反映公司当前的成长性，而未来的成长性仍有较大的不确定性。按照双变 量分组的方法，我们发现部分指标能够提升对下一期业绩增速的预期，包 括单季度 ROE、单季度 ROE 斜率、单季度净利润同比增速斜率、SUE、 单季度营业收入同比增速、单季度经营性现金流净额。 ❑ 综合筛选出的基本面指标构建成长预期组合，区间年化收益接近 26%，夏 普比率 0.95，卡玛比率 0.59。相对比中证 500 指数，超额年化收益接近 19%，从 2012 年以来每年均能战胜中证 500 指数。自 2012 年以来该组合 平均持仓 175 只，平均单边换手约为 32%。 ❑ 进一步地，探究什么样的量价指标可以进一步提升成长预期组合表现。1） 盈余公告次日开盘超额较高的股票组合，其业绩表现是受到市场认可的， 未来业绩表现更好；2）换手率均线标准差较小的组合表现较为优异；3） Amihud 非流动性大的组合业绩表现则较为亮眼。 ❑ 运用上述指标构建技术面精选成长预期组合，回测区间年化收益超过 40%，夏普比率（1.45）与卡玛比率（1.07）均超过 1。相比中证 500 指 数，组合超额年化收益高达 33%。从 2012 年以来每年均能战胜中证 500 指数，且每年超额均在 10%以上。组合平均单边换手约为 67%，各期平均 市值为 113 亿元。</doc> <doc id='22'>资料来源：招商证券、Wind 我们以招商证券量化团队因子库中的成长因子为例，对其有效性进行测试。 表 1 中我们列出了因子回测的框架。回测区间为 2010 年 1 月 1 日至 2025 年 6 月 30 日，每个月最后一个交易日进行调仓，股票权重为等权方式，股票样本池 为全市场，剔除上市不足 180 天、停牌、涨跌停、ST 股票。如无特殊说明，本 文的因子测试均采用此因子回测框架。</doc> <doc id='23'>表 1：因子回测框架 | 项目 | 内容 | | --- | --- | | 回测区间 | 2010.1.1-2025.6.30 | | 调仓频率 | 月度 | | 调仓日 | 最后一个交易日 | | 样本空间 | 全市场 | | 股票筛选 | 剔除上市不足 天、停牌、涨停和 股票 180 ST | | 市值行业中性化 是 | | | IC 测试 IC | 指标为因子值与下一期股票收益率的秩相关系数 | | | 在每个月最后一个交易日后，根据因子值大小将样本空间内的股 | | 分组测试 | 票分成 10 组，每组组内进行等权配置计算各组历史表现。多头 | | | 组为因子值最大的组，空头组为因子值最小的组 | | 基准 | 样本空间内股票的等权组合 | | 交易费率 | 暂不考虑交易费率 | | 资料来源：Wind 资讯、招商证券 | | 如图 4 所示，招商证券量化团队因子库中涵盖单季度净利润同比增速、标准 化预期化盈利、单季度 ROE 同比、净利润增速加速度等指标。除此以外，还有 许多成长因子，正如我们上文所介绍的，但是由于该部分并不是本文研究的重点， 这里我们暂且不做一一展示，感兴趣的投资者欢迎与我们做进一步的交流。</doc> <doc id='24'>图 4：招商证券量化团队因子库成长因子概述 | 因子名称 | 构造方式 | 参考方向 | | --- | --- | --- | | 单李废净利润同比增速 | 单李度归母净利润同比增长率 | 正向 | | 单李度营业收入同比增速 | 单季度营业收入同比增长率 | 正向 | | 单季度营业利润同比增速 | 单率度营业利润同比增长率 | 正向 | | 标准化预期外盈利 | (当前李度归母净利润 -(去年同期单度归母净利润+过去 8个 李度单率归母净利润同比增长均值) / / 过去 8个季度的单 | 正向 | | | 李度归母净利润同比增长值的标准差 | | | 标准化预期外收入 | (当前李度营业收入 - (去年同期单度营业收入+过去 8个 李度单率度营业收入同比增长均值)/过去8个季度的单 | 正向 | | | 李营业收入同比增长值的标准差 | | | 单季度ROE同比 | ROE单季度同比变化 | 正向 | | 单季度ROA同比 | ROA单季度同比变化 | 正向 | | 净利润同比加速度 | 单率度营业利润同比增速的一阶差分 | 正向 | | 净利润TTM环比增速 | 净利润TTM环比增长率 | 正向 | 资料来源：招商证券 表 2 我们具体展示了净利润 TTM 环比增速因子的回测表现。净利润 TTM 环比增速因子在回测区间内的 IC 均值为 2.91%，ICIR 为 0.57，t 值为 7.8。从 分组测试来看，该因子在全 A 市场中分 10 组年化收益单调性一般，但多头组收 益最高，年化收益 15.14%，年化超额 6.28%。</doc> <doc id='25'>表 2：净利润 TTM 环比增速因子回测数据展示 | Rank_IC | Rank_IC 均值 | 胜率（%） | IC_IR | t 统计量 | 最大值 | 最小值 | | --- | --- | --- | --- | --- | --- | --- | | 数据 | 2.91% | 71.51 | 0.57 | 7.80 | 16.79% | -12.50% | | 多空组合 | 年化收益 | 多空卡玛 | 多头年化收益 | 多头年化超额 | 多头夏普 | 多头双边换手 | | | 10.28% | 0.60 | 15.14% | 6.28% | 0.54 | 5.46 | 资料来源：Wind，招商证券；2010/1/1-2025/6/30</doc> <doc id='30'>资料来源：Wind，招商证券；2010/1/1-2025/6/30 资料来源：Wind，招商证券；2010/1/1-2025/6/30 受篇幅限制，其他成长因子的因子测试结果我们就不一一列示，图 7 中我们 统一列出了其他成长因子的表现。综合来看，标准化预期外盈利因子（SUE） 表现较为优异。SUE 因子在回测区间内的 IC 均值为 3.06%，t 值为 6.65，较为 显著；多头组超额年化

1. Report Industry Investment Rating No relevant content provided. 2. Core Viewpoints of the Report - This week, the Hang Seng Index rose, the equity market declined overall, the ChiNext Index fell significantly, and the large - cap value style was dominant. In terms of industries, consumer services and textile and apparel led the gains, while communications, electronics, and computers lagged [1][2][6]. - The average return of the full - market active equity funds was - 0.80%. Funds with better performance were mostly heavy in industries such as pharmaceuticals, electronics, and basic chemicals. Among industry - themed funds, pharmaceutical sector funds had the leading average return, while TMT sector funds had relatively lower average returns [1][2]. - The bond market rose overall this week. The average return of short - term bond funds was 0.04%, and that of medium - and long - term bond funds was 0.07%. Bond funds with equity exposure had an average positive return, and the convertible bond market also rose, with convertible bond funds having an average positive return [1][2]. - As of November 12, 2025, the average returns of low - risk, medium - risk, and high - risk FOF funds in the sample in the past week were 0.32%, 0.58%, and 0.59% respectively [2]. - During the statistical period, the average increases of equity - oriented, index - type, alternative, and bond - type QDII funds were 1.16%, 1.74%, 2.04%, and 0.26% respectively. REITs rose by an average of 0.84% this week [2]. 3. Summary by Relevant Catalogs 3.1 Market Review - The equity market declined overall, with the CSI 300 Index closing at 4628, down 1.08%; the Shanghai Composite Index at 3990, down 0.18%; the Shenzhen Component Index at 13216, down 1.4%; and the ChiNext Index at 3112, down 3.01%. In the Hong Kong stock market, the Hang Seng Index rose 1.26%, and the Hang Seng Tech Index fell 0.42% [6]. - In terms of industries, consumer services and textile and apparel led the gains, with increases of over 4%, while communications, electronics, and computers lagged [8]. 3.2 Key Fund Tracking 3.2.1 Active Equity - **Fund Performance**: The average return of the full - market funds in the sample was - 0.80%. Funds with better performance were heavy in industries such as pharmaceuticals, electronics, and basic chemicals. Among industry - themed funds, pharmaceutical sector funds had the leading average return, while TMT sector funds had relatively lower average returns [13][14]. - **Position Calculation**: The positions of common stock - type and partial - stock hybrid funds both increased this week. Compared with the previous week, the position of common stock - type funds increased by 0.56 percentage points, and that of partial - stock hybrid funds increased by 1.10 percentage points. Actively managed partial - stock funds increased their allocations to growth, consumption, and stability sectors and reduced their allocations to finance and cyclical sectors. In terms of sub - industries, the allocations to electronics, beauty care, and food and beverage increased, while those to automobiles, non - ferrous metals, and basic chemicals decreased [19]. 3.2.2 Bond - type Funds - **Bond Market Performance**: The bond market rose overall this week. The ChinaBond Total Wealth Index closed at 246.41, up 0.07% from last week; the ChinaBond Treasury Bond Index at 246.76, up 0.05%; and the ChinaBond Credit Bond Index at 224.99, up 0.03%. The CSI Non - Pure Bond Fund Index rose 0.21% compared with last Thursday. The CSI Convertible Bond Index closed at 491.71, with a weekly increase of 0.52%, and the trading volume was 349.4 billion yuan, a change of 6.722 billion yuan from last week [23][25]. - **Fund Performance Overview**: The average return of short - term bond funds was 0.04%, and the median was 0.04%; the average return of medium - and long - term bond funds was 0.07%, and the median was 0.06%. The average return of first - tier bond funds was 0.09%, and the median was 0.07%; the average return of second - tier bond funds was 0.04%, and the median was 0.05%. The average return of partial - bond hybrid funds was 0.03%, and the median was 0.05%; the average return of low - position flexible allocation funds was - 0.06%, and the median was - 0.01%. The average return of convertible bond funds was 0.17%, and the median was 0.17% [27][29][32]. 3.2.3 FOF Funds - The average returns of low - risk, medium - risk, and high - risk FOF funds in the sample in the past week were 0.32%, 0.58%, and 0.59% respectively [34]. 3.2.4 QDII Funds - During the statistical period, the average increases of equity - oriented, index - type, alternative, and bond - type QDII funds were 1.16%, 1.74%, 2.04%, and 0.26% respectively [35][36]. 3.2.5 REITs Funds - REITs rose by an average of 0.84% this week. Among them, CICC Liandong Science and Technology Innovation Industrial Park REIT led the gains, rising 6.24% in the past week. Huaxia Hefei High - tech Industrial Park REIT had the strongest liquidity, with a trading volume of 116.2895 million yuan in the past week [37].

Quantitative Models and Construction Methods 1. Model Name: DTW Timing Model - **Model Construction Idea**: The model is based on a similarity approach, examining the similarity between current index trends and historical trends. It selects historical segments with high similarity as references and calculates the weighted average future returns and standard deviations of these segments to generate trading signals[26][28]. - **Model Construction Process**: 1. Use the DTW (Dynamic Time Warping) distance algorithm instead of Euclidean distance to measure similarity, as DTW can better handle time series misalignment issues[28]. 2. Select historical market segments with high similarity to the current market trend. 3. Calculate the weighted average future returns and standard deviations of the selected historical segments, where the weights are the inverse of the DTW distance. 4. Generate trading signals based on the average future returns and standard deviations[26]. 5. The model incorporates improved DTW algorithms, such as those with boundary constraints proposed by Sakoe-Chiba and Itakura, to address issues like "pathological matching" in traditional DTW algorithms[30]. - **Model Evaluation**: The DTW timing model is effective in general market conditions but may face challenges during periods of sudden macroeconomic policy changes, as observed in Q3 2023[17]. 2. Model Name: Foreign Capital Timing Model - **Model Construction Idea**: This model leverages information embedded in the price changes of two offshore assets related to A-shares: FTSE China A50 Index Futures (Singapore market) and the Southbound A50 ETF (Hong Kong market)[35]. - **Model Construction Process**: 1. Construct two indicators using FTSE China A50 Index Futures: basis and price divergence. 2. Combine these indicators to form the FTSE China A50 Index Futures timing signal. 3. Construct a price divergence indicator using the Southbound A50 ETF. 4. Combine the timing signals from the two assets to form the final foreign capital timing signal[35]. --- Model Backtesting Results 1. DTW Timing Model - **Absolute Return**: 38.56% since November 2022[17] - **Maximum Drawdown**: 21.32% since November 2022[17] - **2024 Performance on CSI 300**: - Absolute Return: 36.58% - Maximum Drawdown: 21.36% - Win Rate: 50.00% - Profit-Loss Ratio: 3.39[19] 2. Foreign Capital Timing Model - **Full Sample Performance (2014-2024)**: - Annualized Return: 18.96% (long-short), 14.19% (long-only) - Maximum Drawdown: 25.69% (long-short), 17.27% (long-only) - Daily Win Rate: ~55% - Profit-Loss Ratio: >2.5[21] - **2024 Performance (Out-of-Sample)**: - Absolute Return: 35.42% (long-only) - Maximum Drawdown: 8.23%[24] --- Quantitative Factors and Construction Methods No specific quantitative factors were explicitly mentioned in the report. --- Factor Backtesting Results No specific factor backtesting results were explicitly mentioned in the report.

1. Report Industry Investment Rating No relevant content provided. 2. Core View of the Report The report focuses on the performance and capital flow of the ETF fund market in the past week (from November 10th to November 14th), including the overall market, different popular sub - types, and innovative themes and sub - industries, to provide reference for investors [1]. 3. Summary According to Relevant Catalogs 3.1 ETF Market Overall Performance - **Market Performance**: Most stock ETFs rose this week. Hong Kong stock pharmaceutical and biological ETFs had the largest increase, with an average increase of 6.10% for funds above a certain scale. Conversely, A - share TMT ETFs and A - share growth ETFs declined significantly, with average decreases of 4.37% and 3.20% respectively for funds above a certain scale [2][5]. - **Capital Flow**: Funds flowed into A - share TMT ETFs and commodity ETFs, with net inflows of 8.239 billion yuan and 5.957 billion yuan respectively throughout the week. In contrast, A - share large - cap ETFs experienced capital outflows, with a net outflow of 7.556 billion yuan throughout the week [3][7]. 3.2 Different Popular Sub - type ETF Fund Market Performance - **A - share ETFs**: Include various types such as broad - based index (full - market, large - cap/super large - cap, small - and medium - cap, science and technology/growth enterprise board), industry (TMT, mid - stream manufacturing, consumption, pharmaceutical and biological, cycle, financial and real estate), SmartBeta (value, growth, dividend, free cash flow), and theme. Different types of funds have different performance in terms of weekly capital flow, weekly return, recent one - month return, and year - to - date return [14][15][16]. - **Hong Kong stock ETFs**: Also cover broad - based index, industry (TMT, mid - stream manufacturing, consumption, pharmaceutical and biological, financial and real estate), SmartBeta (dividend), and theme. Each type shows different performance characteristics [29][30][31]. - **Hong Kong - Shanghai - Shenzhen ETFs**: Include industry and theme types, with different performance in terms of capital flow and return [37][38]. - **US stock ETFs**: Divided into broad - based index and industry types, with corresponding performance data [39][40]. - **Other QDII - ETFs (excluding Hong Kong and US stocks)**: Provide performance information of relevant funds [41]. - **Bond ETFs**: Present the performance of bond - related ETFs [42]. - **Commodity ETFs**: Mainly gold - related ETFs, all showing a certain degree of increase this week [43]. 3.3 Innovative Themes and Sub - industry ETF Fund Market Performance - **TMT Innovation Theme**: Indexes such as animation and games, film and television themes, and financial technology have different weekly and year - to - date returns, and the corresponding representative funds also show corresponding performance [45]. - **Consumption Sub - industry**: Indexes like liquor, food and beverage, and tourism have different performance, and the representative funds follow suit [46]. - **Pharmaceutical Sub - industry**: Indexes such as vaccine and biotechnology, traditional Chinese medicine, and innovative drugs have different returns, and the corresponding funds perform accordingly [47]. - **New Energy Theme**: Indexes including power utilities, green power, and photovoltaic industry have different performance, and the representative funds show corresponding trends [48]. - **Central and State - owned Enterprise Theme**: Indexes such as mainland state - owned enterprises, Hong Kong stock central enterprise dividends, and central enterprise innovation have different returns, and the corresponding funds perform accordingly [49][50]. - **Stable Growth Theme**: Indexes such as coal, real estate, and non - ferrous metals have different performance, and the representative funds follow suit [51]. - **Hong Kong - Shanghai - Shenzhen/Hong Kong Stock Connect Sub - industry**: Indexes such as Hong Kong - Shanghai - Shenzhen Internet, Hong Kong securities, and Hong Kong stock connect pharmaceuticals have different returns, and the corresponding funds perform accordingly [52]. - **Dividend/Dividend Low - Volatility Index Family**: Indexes such as CSI 300 Dividend, CSI Dividend Low - Volatility, and SSE Dividend have different performance, and the corresponding funds follow suit [53]. - **Growth Enterprise Board Index Family**: Indexes such as Science and Technology Innovation Chip, Growth Enterprise Board Growth, and Growth Enterprise Board 50 have different performance, and the corresponding funds follow suit [54].

- The report utilizes a random forest model to predict the impact of index constituent stock adjustments on individual stocks' excess returns. The model is designed to handle complex, multi-dimensional, and non-linear problems effectively[13][17][24] - The random forest model selects features based on the logic that passive index funds adjust stock weights following index constituent changes, impacting related stocks. Key features include changes in passive fund holdings, stock liquidity, company market capitalization, and stock price trends[13][15][17] - The construction process of the random forest model involves training on historical data to predict excess returns for stocks affected by index adjustments. The model uses feature selection to enhance generalization ability and focuses on short-term impacts post-announcement[13][17][24] - The evaluation of the random forest model indicates its effectiveness in distinguishing the impact of index adjustments on stocks, particularly in sample-out tests. It successfully identifies stocks with significant excess returns or reduced negative effects[13][17][24] - The backtesting results show that stocks added to the CSI 300 index achieved an average excess return of 2.53% within 10 days post-announcement, while stocks added to the CSI 500 index achieved an average excess return of 1.01% in the same period[17][23][24] - Detailed group performance for stocks added to the CSI 300 index shows excess returns of 2.11% (group_1) and 1.48% (group_5) within 10 days, with a mean return of 2.53%. For the CSI 500 index, group_1 achieved 2.29%, group_5 achieved 0.88%, and the mean return was 1.01% within 10 days[23] - For stocks removed from the indices, the model shows reduced negative effects. CSI 300 stocks in group_1 achieved 1.44% within 10 days, while group_5 showed -0.80%, with a mean return of -0.25%. CSI 500 stocks in group_1 achieved 0.28%, group_5 achieved 0.48%, and the mean return was -0.11% within 10 days[31]