证券研究报告|投资策略周报 [Table_Date] 2026 年 01 月 25 日 [Table_Title] 保持慢牛上涨的趋势不变，聚焦三条配置主线 [Table_Title2] 投资策略周报 [Table_Summary] ·市场回顾：本周全球股指跌多涨少，港股、美股和欧洲股市均呈现下跌，A 股上证指数、深证成指则分别 小幅上涨 0.8%和 1.1%。A 股市场大小盘风格分化，小盘强于大盘，主要宽基指数中，微盘股指、中证 500、 中证 2000 领涨，上证 50、沪深 300 领跌。行业方面，周期和科技成长板块占优，建筑材料、石油石化、钢 铁、化工领涨；大金融、通信、食品饮料领跌。大宗商品方面，贵金属进一步走强，COMEX 白银、COMEX 黄 金价格续创历史新高；国内黑色系商品延续弱势震荡。外汇方面，美元指数失守 98 关口，人民币对美元汇率 继续升值。 ·市场展望：保持慢牛上涨的趋势不变，聚焦三条配置主线。近两周以来，在"逆周期调节"措施下，A 股 主要宽基 ETF 净流出与融资余额高位小幅回落，有效控制了交易节奏；同时，市场成交额仍处相对高位，小 盘成长承接力量持续强劲，春季行情斜率如期放缓 ...

Quantitative Models and Construction Methods Model: AI CSI 1000 Enhanced Portfolio - **Construction Idea**: The model is based on the full-spectrum fusion factor, which integrates both high-frequency and low-frequency price-volume data using deep learning and multi-task learning techniques[6][7] - **Construction Process**: 1. Train 27 high-frequency factors using a deep learning model to obtain high-frequency deep learning factors 2. Use multi-task learning to extract end-to-end features from low-frequency price-volume data, resulting in low-frequency multi-task factors 3. Combine the high-frequency and low-frequency factors to form the full-spectrum fusion factor[6] - **Evaluation**: The model shows significant excess returns and a high information ratio, indicating strong performance and effective risk management[1][7] - **Backtest Results**: - Annualized excess return: 21.60% - Annualized tracking error: 6.06% - Information ratio (IR): 3.57 - Maximum drawdown of excess return: 7.55% - Calmar ratio of excess return: 2.86[1][7] Model: LLM-FADT Text Stock Selection Strategy - **Construction Idea**: The model enhances the BERT-FADT strategy by incorporating additional interpretations from a large language model (LLM), including new title interpretations, market catalysts, implied meanings, potential risks, and return guidance[2][14][17] - **Construction Process**: 1. Input six types of text into a fine-tuned FinBERT model: original text, new title interpretations, market catalysts, implied meanings, potential risks, and return guidance 2. Convert these texts into text feature vectors 3. Train an XGBoost model using these enriched text features[17] - **Evaluation**: The LLM-FADT strategy is more stable and has smaller excess drawdowns compared to the BERT-FADT strategy, showing better performance in extreme market conditions[2][14][20] - **Backtest Results**: - Annualized return: 30.10% - Annualized excess return: 25.52% - Sharpe ratio: 1.18 - Information ratio (IR): 2.00[2][20][24] Model: AI Industry Rotation Model - **Construction Idea**: The model uses the full-spectrum fusion factor to score 32 primary industries and constructs a weekly rebalancing strategy by equally weighting the top 5 industries[3][38] - **Construction Process**: 1. Score each industry using the full-spectrum fusion factor based on the industry component stocks 2. Select the top 5 industries with the highest scores 3. Equally weight these industries and rebalance weekly[38][43] - **Evaluation**: The model effectively utilizes AI's feature extraction capabilities to capture patterns in multi-frequency price-volume data, complementing top-down strategies[3][38] - **Backtest Results**: - Annualized return: 26.87% - Annualized excess return: 19.02% - Maximum drawdown of excess return: 12.43% - Sharpe ratio of excess return: 1.85[3][41] Model: AI Thematic Index Rotation Model - **Construction Idea**: The model scores 133 thematic indices using the full-spectrum fusion factor and constructs a weekly rebalancing strategy by equally weighting the top 10 thematic indices[4][28] - **Construction Process**: 1. Score each thematic index using the full-spectrum fusion factor based on the index component stocks 2. Select the top 10 thematic indices with the highest scores 3. Equally weight these indices and rebalance weekly[28][31] - **Evaluation**: The model leverages AI to identify and capitalize on trends in thematic indices, providing a diversified and dynamic investment approach[4][28] - **Backtest Results**: - Annualized return: 16.92% - Annualized excess return: 9.37% - Maximum drawdown of excess return: 20.79% - Sharpe ratio of excess return: 0.73[4][30] Model Backtest Performance AI CSI 1000 Enhanced Portfolio - Annualized excess return: 21.60% - Annualized tracking error: 6.06% - Information ratio (IR): 3.57 - Maximum drawdown of excess return: 7.55% - Calmar ratio of excess return: 2.86[1][7] LLM-FADT Text Stock Selection Strategy - Annualized return: 30.10% - Annualized excess return: 25.52% - Sharpe ratio: 1.18 - Information ratio (IR): 2.00[2][20][24] AI Industry Rotation Model - Annualized return: 26.87% - Annualized excess return: 19.02% - Maximum drawdown of excess return: 12.43% - Sharpe ratio of excess return: 1.85[3][41] AI Thematic Index Rotation Model - Annualized return: 16.92% - Annualized excess return: 9.37% - Maximum drawdown of excess return: 20.79% - Sharpe ratio of excess return: 0.73[4][30]

Group 1 - Since September 2024, the proportion of southbound capital in transaction volume has rapidly increased to 20%-30%, nearly doubling compared to before 2024, indicating a significant shift in market dynamics [3][8] - Historical reviews of two rounds of competition for pricing power in the Hong Kong stock market occurred in 2016-2017 and 2020-2021, typically initiated by policy optimizations and inflows of incremental capital [15][28] - The current southbound capital inflow is characterized by a higher proportion of medium to long-term funds, with insurance capital making 41 stakes in 2025, 35 of which were in H-shares, marking a record high in the past decade [3][31] Group 2 - The industries where southbound capital and Chinese capital have pricing power include semiconductors and dividend-paying sectors, while industries lacking pricing power include internet, hardware, software services, home appliances, and media [3][36] - The top five industries by southbound capital holdings include coal (41.8%), semiconductors (32.7%), environmental protection (24.5%), oil and petrochemicals (24.1%), and pharmaceutical biology (20.5%) [37] - The active management public funds have a low preference for Hong Kong stocks, with significant holdings concentrated in AI-related CSP giants, electronics, and innovative pharmaceuticals [46] Group 3 - The current sentiment in the Hong Kong stock market has fully reflected negative factors such as US-China trade friction and the high unlock peak at the end of last year, suggesting potential upward investment opportunities if liquidity pressure eases [53][54] - The spring rally in the Hong Kong stock market has a high probability of success, with southbound capital and foreign capital expected to net inflow at the beginning of the year, driven by the demand for core Chinese assets [53][54] - The pricing power of southbound capital is rapidly increasing, with expectations of a potential upward beta in the Hong Kong stock market at the beginning of the year [3][53]

证券研究报告|投资策略周报 [Table_Date] 2026 年 01 月 25 日 [Table_Title] 保持慢牛上涨的趋势不变，聚焦三条配置主线 [Table_Title2] 投资策略周报 [Table_Summary] ·市场回顾：本周全球股指跌多涨少，港股、美股和欧洲股市均呈现下跌，A 股上证指数、深证成指则分别 小幅上涨 0.8%和 1.1%。A 股市场大小盘风格分化，小盘强于大盘，主要宽基指数中，微盘股指、中证 500、 中证 2000 领涨，上证 50、沪深 300 领跌。行业方面，周期和科技成长板块占优，建筑材料、石油石化、钢 铁、化工领涨；大金融、通信、食品饮料领跌。大宗商品方面，贵金属进一步走强，COMEX 白银、COMEX 黄 金价格续创历史新高；国内黑色系商品延续弱势震荡。外汇方面，美元指数失守 98 关口，人民币对美元汇率 继续升值。 ·市场展望：保持慢牛上涨的趋势不变，聚焦三条配置主线。近两周以来，在"逆周期调节"措施下，A 股 主要宽基 ETF 净流出与融资余额高位小幅回落，有效控制了交易节奏；同时，市场成交额仍处相对高位，小 盘成长承接力量持续强劲，春季行情斜率如期放缓 ...

个股方面，本周中国平安、美的集团、海光信息、寒武纪-U、赣锋锂业获主力净流入居前，均超10亿 元；净流出方面，三花智控、中际旭创、香农芯创遭主力净流出居前，分别为61.4亿元、49.88亿元、 41.69亿元。 按申万一级行业来看，本周银行、非银金融、有色金属、煤炭、石油石化五大行业获得主力资金青睐， 其中，银行业获主力净流入47.52亿元；在净流出方面，电子、通信、计算机、电力设备、机械设备行 业均遭抛售超百亿元。 ...

2025 年 12 月 21 日，中国公司治理 50 人论坛（CCG50）在 "中国公司治理 50 人论坛第六届主题论坛（第十九届中国公司治理论坛）" 上发布了 2025 年度 中国上市公司治理和 ESG 系列优秀企业榜单。该系列榜单由 CCG50 委托北京师范大学公司治理与企业发展研究中心高明华教授团队开发，评价样本涵盖截 至 2025 年 4 月 30 日上市满一年，且剔除退市、拟退市及未发布年报公司后的 5292 家上市公司，最终解释权归属 CCG50，联系人曹向东（电话 15901348949，同微信号）。 | 2025 年度中国上市公司治理百佳企业榜单(总榜) | | --- | | 2025 年度中国上市公司中小投资者权益保护百佳企业榜单 4 | | 2025 年度中国上市公司董事会治理百佳企业榜单 . | | 2025 年度中国上市公司企业家(总经理)百佳榜单 | | 2025 年度中国上市公司财务治理百佳企业榜单 | | 2025 年度中国上市公司自愿性信息披露百佳企业榜单 17 | | --- | | 2025 年度中国出海上市公司治理五十佳企业榜单 . | | 2025 年度中国非金融业上 ...

- The report introduces a quantitative model based on Convolutional Neural Networks (CNNs) to analyze price-volume data and predict future prices. The model standardizes price-volume data into graphical representations and maps learned features to industry theme indices, such as the CSI Petrochemical Industry Index, CSI Subdivision Chemical Industry Theme Index, CSI Machine Tool Index, CSI Semiconductor Material Equipment Theme Index, and CSI Nonferrous Metals Index[78][80][81] - The construction process of the CNN model involves transforming individual stock price-volume data within a specific window into standardized graphical charts. These charts are then input into the CNN for feature extraction and prediction modeling. The learned features are subsequently applied to identify and allocate industry themes[78][80] - The evaluation of the CNN model highlights its ability to capture complex patterns in price-volume data and effectively map these patterns to industry themes. This approach provides a novel perspective for quantitative investment strategies[78][81] - Backtesting results indicate that the CNN model's latest configuration suggests a focus on themes such as petrochemicals, chemicals, machine tools, semiconductors, and nonferrous metals. Specific indices include the CSI Petrochemical Industry Index, CSI Subdivision Chemical Industry Theme Index, CSI Machine Tool Index, CSI Semiconductor Material Equipment Theme Index, and CSI Nonferrous Metals Index[80][81]

Group 1 - The core viewpoint emphasizes that the "steady and far-reaching" strategy is facilitating the transition of the spring market into subsequent phases without disrupting the established path of the spring market [1][2] - The spring market is characterized by incremental games and favorable conditions for long positions, supported by factors such as increased equity allocation by residents and active trading funds [2][3] - Short-term, the focus is on discovering bottom assets, with cyclical Alpha investments expanding towards more cyclical turning points, while the overall profit effect is nearing a high [1][4] Group 2 - The spring market is positioned as an extension and expansion phase of the high valuation area of the 2025 technology structural market, with expectations of a consolidation phase following the spring market [3] - Short-term, cyclical Alpha is becoming a key focus for market participants looking to exploit low positions, with notable sectors including commercial aerospace and AI applications showing rebound opportunities [4][7] - The long-term outlook remains positive for sectors such as technology and cyclical Alpha, with a focus on areas like overseas computing chains and AI applications [7][12]

来源：渤海证券股份有限公司 投资要点： 市场回顾，近5 个交易日（1 月16 日-1 月22 日），重要指数涨跌不一；其中，上证综指收涨0.24%，创 业板指收跌1.17%；风格层面，沪深300收跌0.58%，中证500 收涨2.00%。成交方面显著缩量，两市统计 区间内成交13.80 万亿元，日均成交额降至2.76 万亿元，较前五个交易日的日均成交额减少6834.69 亿 元。行业方面，申万一级行业涨多跌少，其中，建筑材料、石油石化、基础化工行业涨幅居前，而传 媒、计算机、银行行业跌幅居前。 数据方面，国家统计局公布12 月经济数据。投资端，2025 年固定资产投资同比下降3.8%，较2025 年1- 11 月回落1.2 个百分点，投资三项继续走弱。消费端，2025 年12 月社会消费品零售总额同比增速也进 一步放缓，主要受政策退坡和高基数影响。整体而言，2025 年全年经济增速呈现前高后低，全年目标 顺利完成。展望来看，2026 年年初，出口端在结构优化、产品竞争优势以及经贸环境改善的支撑下将 继续成为支撑项；与此同时，年初新一轮政策部署落地，保证了政策的接续性，有望为经济"开门红"提 供助力。 政策方面 ...

每经记者｜刘明涛 每经编辑｜肖芮冬 成分股方面，TCL电子本周涨幅高达24.77%；潍柴动力继续保持强势，周涨幅也达到12.33%；此外，百度集团-SW、中国重汽、招商蛇口、四川长虹、保利 发展、中国海油、中国石化的周涨幅也均超过5%。周市值增长方面，中国海油本周市值增长达到983.87亿元，市值增长接近千亿元。 | | | 有业内人士指出，随着浙江威乐公司在技术、运营能力上的持续提升，以及业务规模效应逐步显现，该业务的盈利能力和利润率水平有望随之稳步改善。 与此同时，四川长虹一直强化AI技术赋能，其在AI技术领域的聚焦突破，不仅有效提升了居家空气品质，也持续释放新的消费潜力。日前，在2026年国际 消费电子展上，四川长虹与高通联合推出新款AT TV Q10Light。据悉，这款长虹新品将搭载端侧AI字幕翻译大模型，可实现影视内容的音频对话到指定语种 的实时转译与字幕同步显示。1月6日，长虹空调还发布了AI人感空调系列矩阵新品，全面覆盖不同场景下的用户需求，将技术优势转化为场景化体验优 势。 在之前一次机构调研中，四川长虹就曾表示，长虹云帆AI大模型核心聚焦智慧家电领域，是家电领域首个通过国家备案的大模型。优 ...