[Table_Page] 金融工程|专题报告 2026 年 1 月 5 日 证券研究报告 [Table_Title] 金融工程：大类资产配置分析月报 （2025 年 12 月） PMI 回升至荣枯线以上，当前看多权益资产 [Table_Summary] 报告摘要： 表：大类资产最新观点(20251231) 数据来源：Wind, 广发证券发展研究中心 | [分析师： Table_Author]李豪 | | --- | | SAC 执证号：S0260518070001 | | 021-38003569 | | lhao@gf.com.cn | | 分析师： 安宁宁 | | SAC 执证号：S0260512020003 | | SFC CE No. BNW179 | | 0755-23948352 | | anningning@gf.com.cn | | 请注意，李豪并非香港证券及期货事务监察委员会的注册 | 持牌人，不可在香港从事受监管活动。 972918116公共联系人2026-01-05 14:56:19 1 / 15 识别风险，发现价值 请务必阅读末页的免责声明 图：大类资产配置组合表现 0 1 2 3 4 ...

Quantitative Models and Construction Methods 1. **Model Name**: Dividend Style Timing Model **Model Construction Idea**: The model is designed to time the dividend style by aggregating multiple factors that influence the performance of dividend-related stocks. The comprehensive factor value is used to determine the timing signal for the dividend style[6]. **Model Construction Process**: - The model aggregates several sub-factors, including U.S. Treasury yields, the spread between dividend yield and Chinese bond yields, and industry sentiment indicators. - The comprehensive factor value is calculated as a weighted sum of these sub-factors. - The model outputs a signal based on whether the comprehensive factor value is greater than or less than zero. A positive value indicates a favorable timing signal, while a negative value suggests an unfavorable signal[6][7]. **Model Evaluation**: The model provides a systematic approach to timing the dividend style, but its effectiveness depends on the stability and predictive power of the underlying factors[6][7]. --- Model Backtesting Results 1. **Dividend Style Timing Model**: - Comprehensive factor value for the week of 2025.12.22 to 2025.12.26: -0.55 - Comprehensive factor value for the previous week (2025.12.15 to 2025.12.19): -0.72 - The model's comprehensive factor value showed improvement but remained below zero, indicating no positive timing signal[6][7]. --- Quantitative Factors and Construction Methods 1. **Factor Name**: U.S. 10-Year Treasury Yield **Factor Construction Idea**: This factor reflects the impact of U.S. Treasury yields on dividend style performance. A decline in yields is generally considered supportive of dividend stocks[7]. **Factor Construction Process**: The factor value is derived from the weekly change in the 10-year U.S. Treasury yield. A negative value indicates a decline in yields, which is expected to positively influence the dividend style[11]. 2. **Factor Name**: Spread Between Dividend Yield and 10-Year Chinese Bond Yield **Factor Construction Idea**: This factor measures the relative attractiveness of dividend yields compared to risk-free bond yields in China. A wider spread is considered favorable for dividend stocks[7]. **Factor Construction Process**: - The factor value is calculated as the difference between the dividend yield of the CSI Dividend Index and the 10-year Chinese bond yield. - A positive value indicates that dividend yields are higher than bond yields, which is supportive of the dividend style[11]. 3. **Factor Name**: Industry Sentiment Indicator **Factor Construction Idea**: This factor captures the overall sentiment in the industry, which can influence the performance of dividend stocks. Positive sentiment is expected to support the dividend style[7]. **Factor Construction Process**: The factor value is derived from analysts' assessments of industry conditions. A higher value indicates stronger sentiment, which is favorable for dividend stocks[11]. --- Factor Backtesting Results 1. **U.S. 10-Year Treasury Yield**: - Factor value on 2025.12.26: -0.91 - Factor value on 2025.12.19: -1.08 - Factor value on 2025.11.30: -1.32[11] 2. **Spread Between Dividend Yield and 10-Year Chinese Bond Yield**: - Factor value on 2025.12.26: 0.72 - Factor value on 2025.12.19: 0.32 - Factor value on 2025.11.30: -0.32[11] 3. **Industry Sentiment Indicator**: - Factor value on 2025.12.26: 1.65 - Factor value on 2025.12.19: 1.77 - Factor value on 2025.11.30: 1.97[11]

期指长周期因子小幅下降 金融工程周报 操作评级 股指 ☆☆☆ 国债 ☆☆☆ 王锴 金融工程组 010-58747784 gtaxinstitute@essence.com. cn Z0016943 F03091361 本报告版权属于国投期货有限公司 1 不可作为投资依据，转载请注明出处 2025年12月29日 周度报告 p 截至12月26日当周，期指分化，IH2601上升1.45%， IF2601上升2.79%，IC2601上升4.86%，IM2601上升 4.97%。板块层面，交投情绪升温带动科技与反内卷主线再 度活跃，卫星通信、新能源等板块表现强势。展望后市，当 前市场在资金情绪推动下持续修复，主要宽基指数已接近前 期高位。 p 从高频宏观基本面因子评分来看，期指方面，通胀指标8分， 流动性指标9分，估值指标11分，市场情绪指标9分。期债 方面，通胀指标8分，流动性指标9分，市场情绪指标5分。 期限结构方面，IH、IF、IC、IM 期末持仓量加权年化基差 率（分红调整）分别为 1.05%、-1.36%、-3.5%、- 6.52%，远月合约贴水较上周有所收窄。 p 金融衍生品量化CTA策略上周净值上升0.9 ...

贵金属ETF收益表现偏强 金融工程周报 基金市场回顾： 操作评级 中信五风格-消费★☆☆ 金融工程组 张婧婕 Z0022617 010-58747784 gtaxinstitute@essence.com.cn 权益市场风格 本报告版权属于国投期货有限公司 1 不可作为投资依据，转载请注明出处 2025年12月29日 周度报告 截至2025/12/26当周，通联全A（沪深京）、中证综合债与南华 商品指数周度涨跌幅分别为2.73%、0.07%、4.00%。 公募基金市场方面，近一周权益策略收益反弹，增强指数策略上 涨2.82%；纯债策略收益分化，中长期纯债收益优于短期纯债； 商 品 类 ETF 走 强 ， 白 银 ETF 净 值 大 幅 上 行 ， 周 度 收 益 率 为 17.43%。 中信五风格方面，上周成长与周期风格表现偏强，消费风格小幅 收跌；风格轮动图显示近期稳定风格相对强弱边际回落，消费风 格相对强弱动量环比上升。公募基金池方面，近一周金融与成长 基金超额表现较优，从基金风格系数走势来看市场对成长与消费 风格偏移度小幅回升；本周拥挤度指标相比上周环比提高，周期 风格基金拥挤度回落至历史低分位区间，成 ...

丨证券研究报告丨 金融工程丨点评报告 [Table_Title] 2025 年有效选股因子 报告要点 [Table_Summary] 本文主要回顾 2025 年选股因子在全市场的表现情况。 分析师及联系人 [Table_Author] 郑起 覃川桃 SAC：S0490520060001 SAC：S0490513030001 SFC：BUT353 请阅读最后评级说明和重要声明 %% %% %% %% research.95579.com 1 [Table_Title2] 2025 年有效选股因子 [Table_Summary2] 2025 年因子选股全市场范围内表现较好 收益能力上看，2025 年全市场范围内选股以成交笔数、流动性、拥挤度、价格稳定和反转为代 表的量价因子选股能力更强，成长因子有一定盈利能力，但在失效区间回撤较大。 收益来源上看，分为量价和成长两大类，量价内又以价格稳定、反转为两类主要代表。 2025 年因子收益有着较为明显的区间特征 2025 年 8 月（不含）以前是所有因子的主要收益区间，8 月为因子的第一次分化，表现为量价 因子（除反转）回撤，成长因子收益正常，9 月为所有因子回撤区间，1 ...

本报告版权属于国投期货有限公司 1 不可作为投资依据，转载请注明出处 2025年12月15日 周度报告 期指长周期因子下降 金融工程周报 操作评级 股指 ☆☆☆ 国债 ☆☆☆ 王锴 金融工程组 010-58747784 gtaxinstitute@essence.com. cn Z0016943 F03091361 p 截至12月12日当周，A股整体呈结构化震荡走势，全市场日 均成交额为195万亿元，较上周增加近2600亿元，市场成 交活跃度小幅回升。三大指数涨跌不一，其中上证指数下跌 0.34%。短期增量政策与经济数据信息相对有限，市场结构 性特征显现。 p 从高频宏观基本面因子评分来看，期指方面，通胀指标8分， 流动性指标9分，估值指标11分，市场情绪指标9分。期债 方面，通胀指标8分，流动性指标10分，市场情绪指标6分。 期限结构方面，IH、IF、IC、IM 期末持仓量加权年化基差 率（分红调整）分别为 0.33%、-2.32%、-4.16%、- 9.95%，远月合约贴水再度扩大。 p 金融衍生品量化CTA策略上周净值没有变化。长周期方面， 社融数据虽然小幅超预期，但信贷数据M1和M2等均表现出 了超季 ...

[Table_Page] 金融工程|定期报告 2025 年 12 月 14 日 证券研究报告 | [Table_Title] 金融工程：AI | | | | 识图关注通信、人工 | | | | | --- | --- | --- | --- | --- | --- | --- | --- | | 智能 | | | | | | | | | | | | A | | 股量化择时研究报告 | | | | ] [Table_Summary 市场回顾（本期是指 | | 2025 年 | 12 月 | 8 日—2025 | 年 12 月 | 12 日） | | | 中证 结构表现 | | 沪深 | 中证 | 中证 | 中证 | | 国证 | | (涨幅) | 100 | 300 | 500 | 800 | 1000 | | 2000 | | -0.13% | | -0.08% | 1.01% | 0.21% | 0.39% | | 0.69% | | [分析师： Table_Author]安宁宁 | | | --- | --- | | | SAC 执证号：S0260512020003 | | SFC CE No. BNW1 ...

Core Viewpoint - The article discusses the complex and potentially catastrophic financial arrangements within the AI industry, drawing parallels to the financial crisis of 2008, highlighting the risks associated with high levels of debt and interlinked financial structures among major tech companies [5][6][10]. Company Analysis - CoreWeave, a relatively unknown company, has emerged as a significant player in the AI sector, achieving the largest tech IPO since 2021 and doubling its stock price. It has secured major contracts worth $220 billion with OpenAI, $140 billion with Meta, and $60 billion with Nvidia [5][6]. - Despite its impressive contracts, CoreWeave operates at a loss, projecting $5 billion in revenue against $20 billion in expenses for the year. The company has accumulated $14 billion in debt, with over half due within a year, and faces $34 billion in lease obligations by 2028 [6][7]. Financial Structures - The financial model of CoreWeave relies heavily on a few key clients, with Microsoft accounting for 70% of its revenue, and Nvidia and OpenAI contributing an additional 20%. This creates a precarious dependency on a limited customer base [7]. - The AI industry's financialization is driven by the high costs of infrastructure needed for AI systems, with data center spending expected to exceed $400 billion this year and potentially reach $7 trillion by 2030. Creative financing methods are essential to support these investments [8][10]. Interconnectedness and Risks - Major companies like Nvidia, OpenAI, and others are forming intricate financial relationships, often involving equity stakes in exchange for future profits, which obscures the true financial health of these companies [8][9]. - The article warns that if the anticipated AI revolution does not materialize as expected, the financial ties binding these companies could lead to widespread economic repercussions, potentially more severe than the dot-com bubble burst [10][11]. Debt and Financial Instruments - The AI sector is accumulating significant debt, with estimates suggesting it could reach $1.5 trillion by 2028. This high leverage poses risks to the broader financial system if defaults occur [11][14]. - Companies are utilizing complex financial instruments, such as special purpose vehicles (SPVs) and asset-backed securities, to obscure debt levels and manage financing, reminiscent of practices leading up to the 2008 crisis [12][13]. Regulatory Environment - The article highlights concerns over the lack of regulatory oversight for private equity firms involved in AI financing, which could exacerbate risks in the event of a market downturn. The interconnectedness of private credit and traditional financial institutions raises alarms about potential systemic risks [14][15]. - Recent regulatory rollbacks may expose a broader public to the risks associated with AI financing, contrasting with the more reactive approach taken during the 2008 crisis [15][16].

Market Overview - Overall asset performance shows commodities outperforming equities, with the Korean stock market leading gains[4] - MSCI global index increased by 0.6%, but growth momentum has significantly slowed compared to previous weeks[5] - The yield curve for Chinese bonds is steepening, indicating a "bear steepening" trend, while U.S. bonds are experiencing a "bull steepening" trend[6] Commodity and Currency Trends - 10 out of 13 major commodities recorded price increases, with COMEX silver rising by 101.9% year-to-date[7] - The U.S. dollar index fell by 0.5%, with the euro and pound appreciating by 0.4% and 0.8% respectively; the dollar has depreciated by 8.8% since the beginning of the year[7] Consumer and Industrial Insights - Service consumption has improved year-on-year, with Shanghai Disneyland's visitor index up by 75% compared to last year[10] - Real estate transactions in major cities have seen significant declines, with new home sales down by 32.5% year-on-year[30] Financial Sector Developments - As of November 2025, the total net asset value of public funds reached 36 trillion yuan, with equity funds increasing by 1.55%[24] - The performance evaluation of the investment banking sector is shifting towards enhancing investor experience[23] Company-Specific Highlights - Traffic Bank's net profit growth is projected at 2.3% for 2025, with a target price adjustment to 8.58 yuan based on a 0.72x PB valuation[34] - Didi's EBITA is expected to be 46.0 billion yuan in 2025, with a target market value of 234.7 billion yuan[39]

Group 1 - The company is recruiting interns for positions in Shenzhen, Shanghai, and Beijing, requiring in-person internships with a minimum commitment of three days per week for at least three months [1] - The application deadline for submitting resumes is December 31, 2025 [1] - Interns with outstanding performance may have the opportunity for full-time employment after the internship [1] Group 2 - Responsibilities include data processing, analysis, and assisting researchers with quantitative investment projects [2] - Interns will also assist in the development and tracking of financial engineering strategy models [2] - Additional tasks may be assigned by the team [2] Group 3 - Basic requirements include being a master's or doctoral student in STEM fields or financial engineering, with a strong preference for exceptional fourth-year students [3] - Proficiency in programming languages such as Python and familiarity with SQL databases are essential [3] - Candidates should possess strong self-motivation, analytical skills, and effective communication abilities [3] Group 4 - Preferred qualifications include a solid foundation in financial markets, familiarity with key concepts in stocks, bonds, futures, indices, and funds [4] - A strong mathematical background, research project experience, and published academic papers in SCI or EI are advantageous [4] - Familiarity with financial terminals like Wind, Bloomberg, and Tianruan, as well as knowledge of machine learning and deep learning, is a plus [4] Group 5 - Interested candidates should submit their resumes in PDF format to the specified email address, following a specific naming convention for the email subject [5] - Resumes not adhering to the naming format will be treated as spam [5] - Qualified candidates will be contacted for written tests and interviews after the resume collection deadline [5]