（点击↑↑ 上图查看详情 ） 上周，A股核心指数迎来普涨行情，周五沪指盘中触及3700关键点位， 然而与之相反的则是指数增强策略产品大幅跑输基准指数表现。根据 私募排排网数据显示，沪深300指增、中证500指增、中证1000指增、量化多头（相对中证800）产品在上周的超额涨跌幅分别 为-0.49%、-1.09%、-1.26%、-1.16%，而市场中性产品同样在上周录得-0.83%的平均回撤。（可参考： 最新量化多头私募公司榜揭晓！鸣 石、黑翼、稳博位居前3！ ） 数据来源：同花顺iFinD，截至日期为2025年8月11日至2025年8月15日 图2：上周量化管理人指数成分股跑赢基准指数统计 图1：上周量化管理人基础选股池跑赢基准指数统计 为何上周再现beta涨alpha跌的怪象？ 实际上本次市场行情结构与2024年春节前一周以及924行情首周较为类似，即宽基指数表现强劲，但落 实到全市场选股层面则个股表现相对较弱，并且各版块风格在市场情绪高涨时会容易加速轮动，因此超额获取难度较前期增加。我们将量化 管理人基础选股池（无次新股、无ST股、无*ST股）个股在上周收益做统计后发现，五个交易日中跑赢对标指数的比例均小 ...

指增组合本周超额回撤 分析师 缪铃凯 SAC 证书编号：S0160525060003 miaolk@ctsec.com 相关报告 1. 《沪深 300 增强超额收益创年内新高》 2025-08-09 2. 《 指 增 组 合 本 周 抗 跌 效 果 显 著 》 2025-08-02 3. 《深度学习因子选股体系》 2025- 08-01 证券研究报告 量化选股策略周报/ 2025.08.16 核心观点 ❖ 风险提示：因子失效风险，模型失效风险，市场风格变动风险。 请阅读最后一页的重要声明！ ❖ 本周市场指数表现：截至 2025-08-15，本周上证指数上涨 1.70%，深证 成指上涨 4.55%，沪深 300 上涨 2.37%，上证指数创 2022 年以来新高。 ❖ 我们基于深度学习框架构建 alpha 和风险模型，打造 AI 体系下的低 频指数增强策略，组合周度调仓，年单边换手率约 5.5 倍。最终，通 过组合优化勾连深度学习 alpha 信号与风险信号构建沪深 300、中证 500 和中证 1000 指数增强组合。 ❖ 截至 2025-08-15，今年以来沪深 300 指数上涨 6.8%，沪深 300 指 ...

金 融 工 程 再论沪深 300 增强：从增强组合成分股 内外收益分解说起 本报告导读： 结合适合沪深 300 指数成分股的多因子模型，辅以小市值高增长组合作为成分股外 卫星策略，可较为稳定改善沪深 300 增强策略业绩表现。域内 30%、域外 10%卫星 配置比例下，2016 年以来 300 增强策略年化超额收益 12.6%，跟踪误差 5.2%。 投资要点： 成分股外类似于提供收益弹性的卫星策略，也可采用小市值高增 长、或 GARP 策略等具有类似特征的主动量化组合替代。我们以小 市值高增长组合策略作为域外卫星策略，域内则结合适合沪深 300 指数成分股的多因子模型，可较为稳定改善沪深 300 指数增强策略 业绩表现。至于具体的卫星配置比例，则取决于收益风险偏好和需 求。域内 30%、域外 10%卫星配置比例下，2016 年以来沪深 300 增 强策略年化超额收益 12.6%，跟踪误差 5.2%，信息比 2.38。最极端 的情况，域内、域外都完全用卫星策略替代，则沪深 300 增强策略 年化超额收益 17.5%；但同时跟踪误差也最高，达 7.6%，信息比 2.20，相对回撤 9.0%。 风险提示。本文根据客 ...

Core Viewpoint - The market is experiencing a significant increase in risk appetite, with a notable rise in small-cap growth stocks, supported by a high level of margin financing that has returned to over 2 trillion yuan for the first time since 2015, although its market share has decreased [1][3][5]. Group 1: Market Performance - The three major indices opened higher on August 12, with the Shanghai Composite Index continuing to strengthen after reaching a new high since 2022 [1]. - As of August 11, the margin financing balance has remained above 2 trillion yuan for four consecutive trading days, marking a significant return to this level [1][3]. Group 2: Margin Financing Insights - Despite the high absolute value of margin financing, its proportion of the A-share market's circulating market value is only 2.30%, less than half of the 4.73% seen in April 2015, indicating a healthier leverage structure [3]. - This situation suggests that there may still be marginal incremental capital supporting the market, with leverage levels not yet reaching alarm thresholds [3]. Group 3: Investment Strategy - The current market dynamics favor small-cap growth stocks, which are particularly sensitive to changes in liquidity and risk appetite, as evidenced by the performance of the CSI 1000 index, which has risen 16.55% year-to-date compared to the 8.82% increase of the Shanghai Composite Index [5]. - The 1000 ETF Enhanced (159680) has seen a year-to-date net value increase of 25.25%, significantly outperforming its benchmark, with a substantial inflow of 213 million shares over the year [8]. - The strategy of using index enhancement may be beneficial for investors looking to balance their portfolios while capturing both beta and alpha returns in the small-cap sector [10]. Group 4: Future Considerations - The market's increasing heat and the gains across various sectors suggest that a refined approach to portfolio rebalancing may be prudent, particularly through low-cost dollar-cost averaging or phased entry strategies [11].

Group 1 - The fund is named "E Fund Shanghai Stock Exchange Science and Technology Innovation Board Comprehensive Enhanced Strategy ETF" and is a type of open-ended index fund [23] - The fund will be available for subscription from August 18 to August 22, 2025, with both online and offline cash subscription options [2][23] - The maximum fundraising scale for the fund is set at 2 billion RMB, excluding interest and subscription fees [5][23] Group 2 - Investors must have a Shanghai Stock Exchange A-share account or a securities investment fund account to subscribe to the fund [3][44] - The subscription fee for the fund will not exceed 0.80% of the subscribed amount [4][28] - The fund's investment objective is to pursue returns that exceed the performance benchmark while controlling the average tracking deviation and annualized tracking error [24][25] Group 3 - The fund's underlying index is the Shanghai Stock Exchange Science and Technology Innovation Board Comprehensive Index, which includes stocks listed on the Science and Technology Innovation Board [12][13] - The fund will be managed by E Fund Management Co., Ltd., with Ping An Bank as the custodian [1][64] - The fund's shares will be issued at an initial value of 1 RMB per share [23]

Core Insights - The report emphasizes the development of a deep learning factor selection system for stock prediction and portfolio optimization, shifting from traditional logic-driven methods to data-driven approaches [7][10]. - The system integrates diverse data sources, including daily and minute market data, to enhance the performance of alpha signals [7][10]. - The report outlines the construction of multiple models that utilize different network architectures to extract unique alpha signals, demonstrating low correlation among them [8][54]. Data and Network - The input data consists of three categories: daily market data, minute market data, and manually crafted features, with neural networks independently extracting alpha features from each dataset [11]. - The report describes the use of Long Short-Term Memory (LSTM) networks combined with self-attention mechanisms to capture long-term dependencies in time series data [19]. - A Graph Attention Network (GAT) is employed to model the complex relationships between stocks, providing a global analysis perspective [20]. Alpha Models - The report presents various alpha models, including simple equal-weight, tree model weighting, and network weighting, with a focus on combining multiple signals to enhance robustness [3][3.1][3.2]. - The average Information Coefficient (IC) for the combined factors since 2019 is reported as 11.3% for 5-day IC and 12.4% for 10-day IC, indicating strong predictive power [31][32]. Risk Models - The report highlights the use of neural networks to identify high-dimensional non-linear risk patterns directly from raw price and volume data, enhancing risk control in portfolio construction [9]. Index Enhancement Strategies - The report details the performance of enhanced index strategies based on deep learning alpha signals, with annualized returns reported as follows: - CSI 300 enhanced portfolio: 18.2% annualized return, 14.2% excess return over the index [3][5.1]. - CSI 500 enhanced portfolio: 22.4% annualized return, 17.2% excess return over the index [3][5.2]. - CSI 1000 enhanced portfolio: 29.8% annualized return, 24.5% excess return over the index [3][5.3].

Core Viewpoint - The report demonstrates the effectiveness of deep learning models, specifically GRU and TCN, in stock selection, with GRU showing slightly better performance than TCN+GRU and TCN. The 10-day return prediction model outperforms the 5-day model. The deep learning factors are highly correlated with low volatility and low liquidity factors, indicating potential investment strategies [1][2]. Group 1: Model Performance - The GRU model is confirmed to be effective, with advantages in prediction accuracy and training speed, making it widely used in the industry [1]. - The TCN model, based on CNN architecture, effectively captures long-term dependencies in time series data through causal convolution and residual connections [1]. - The annualized excess returns since 2017 for various indices are as follows: - CSI 300: 11.8% - CSI 500: 13.6% - CSI 1000: 21.7% - CSI 2000: 27.1% The current year's excess returns are -0.4%, 2.7%, 9.9%, and 9.3% respectively [1][3]. Group 2: Single Factor Stock Selection - The single-factor stock selection shows better performance in small and mid-cap stock pools (CSI 1000, CSI 2000), with minimal impact from market capitalization and industry neutrality [2]. - The original factor values in CSI 300 outperform the market capitalization and industry-neutralized factor values, indicating that deep learning factors capture style and industry rotation patterns [2]. Group 3: Composite Factor Stock Selection - Composite factors, when equally weighted, outperform single factors, and the report outlines the construction of index-enhanced strategies with specific constraints on stock turnover and market exposure [3]. - The maximum drawdown for the CSI 300 index-enhanced strategy since January 2017 is -6.0%, with a current year excess return of -0.4% [3]. - Allowing for slight market and industry exposure results in annualized excess returns of 8.8% for CSI 300 and 14.6% for CSI 500, with current year excess returns of -1.7% and 5.2% respectively [3].

Core Viewpoint - The expansion of personal pension index fund Y shares marks a significant development in the investment landscape, with an increase in the number of available funds and a growing interest in index-enhanced products as a key component of pension asset allocation [1][2][5]. Group 1: Fund Expansion and Market Dynamics - On July 28, Huaxia Fund announced the addition of Y shares for its Huaxia ChiNext ETF linked fund to meet the investment needs of personal pension investors, effective from July 31 [1]. - Following this, four index-enhanced funds also announced the establishment of their Y shares, bringing the total number of index funds available for personal pensions to 90, with the overall number of personal pension Y share funds exceeding 300 for the first time [2]. - The approval of new Y shares for index-enhanced funds indicates a growing trend in the market, with more products expected to be submitted for inclusion in the personal pension product directory [3]. Group 2: Fund Management and Performance - Both Guotai Haitong Asset Management and Bodao Fund have introduced their first index-enhanced personal pension funds, showcasing their unique strategies in the competitive landscape [3][4]. - As of July 29, the total number of index-enhanced funds in the market reached 760, with a combined scale of 222.16 billion yuan, highlighting the increasing popularity of these products [4]. - The average annual return for 20 index-enhanced Y shares in the first half of 2025 was 10%, outperforming their performance benchmark by 3.2% [8]. Group 3: Investment Strategy and Market Outlook - Index-enhanced products are becoming a favored choice for pension allocation, as they provide a means to combat inflation and capitalize on economic growth [5]. - The core logic of index-enhanced products is to optimize stock structure to achieve excess returns while adhering to index styles, which is particularly relevant in the current market environment characterized by liquidity-driven growth and structural rotation [6][7]. - The average return of personal pension funds established before 2025 was 6.98%, with Y share index funds achieving an average return of 8.32%, indicating strong performance in the pension fund sector [6].

Group 1 - The core viewpoint of the article is to track the performance of various index enhancement portfolios and the factors influencing stock selection across different indices, highlighting the excess returns achieved by these portfolios [1][2][3]. Group 2 - The performance of the HuShen 300 index enhancement portfolio this week showed an excess return of 0.78%, with a year-to-date excess return of 9.31% [5]. - The performance of the Zhongzheng 500 index enhancement portfolio this week showed an excess return of -0.52%, with a year-to-date excess return of 9.90% [5]. - The Zhongzheng 1000 index enhancement portfolio had an excess return of 0.07% this week, with a year-to-date excess return of 15.69% [5]. - The Zhongzheng A500 index enhancement portfolio reported an excess return of 0.26% this week, with a year-to-date excess return of 9.96% [5]. Group 3 - In the HuShen 300 component stocks, factors such as specificity, EPTTM one-year quantile, and quarterly net profit year-on-year growth performed well [8]. - In the Zhongzheng 500 component stocks, factors like three-month volatility, EPTTM one-year quantile, and expected BP showed good performance [8]. - For Zhongzheng 1000 component stocks, factors such as three-month institutional coverage, three-month reversal, and expected BP performed well [8]. - In the Zhongzheng A500 index component stocks, factors like specificity, three-month reversal, and expected net profit month-on-month growth performed well [8]. Group 4 - The HuShen 300 index enhancement products had a maximum excess return of 1.28%, a minimum of -0.98%, and a median of 0.12% this week [21]. - The Zhongzheng 500 index enhancement products had a maximum excess return of 1.41%, a minimum of -1.31%, and a median of 0.04% this week [21]. - The Zhongzheng 1000 index enhancement products had a maximum excess return of 0.82%, a minimum of -0.47%, and a median of 0.15% this week [21]. - The Zhongzheng A500 index enhancement products had a maximum excess return of 1.16%, a minimum of -0.57%, and a median of -0.04% this week [21].

Core Viewpoint - The A-share market has attracted significant attention due to its strong performance, with the Shanghai Composite Index breaking through key levels, leading to discussions about market entry and potential risks of "catching the falling knife" [2] Group 1: Market Performance and Strategy - The Shanghai Composite Index successfully broke through 3500 points on July 9 and again surpassed 3600 points on July 23, indicating a strong market trend [2] - Quantitative long strategies have shown impressive results in the first half of the year, with an average return of 17.32% for quantitative long strategies, leading among 16 secondary strategies [3] - The average return for the small-cap index enhancement strategy, represented by the CSI 1000 index, reached 20.26%, making it one of the top-performing products [2][3] Group 2: Expert Insights on Investment Strategies - Experts from 15 well-known private equity firms provided insights on whether to invest in quantitative index enhancement products at the current market levels [2] - Montrose Investment believes that the small-cap index enhancement configuration window is still open but requires more refined factor design and position optimization [5] - Inno Asset suggests that the second half of the year may bring new opportunities for factor exploration and strategy adjustment due to a stable macroeconomic environment [10] Group 3: Factors Driving Performance - The strong performance of quantitative index enhancement strategies in the first half of the year is attributed to three main factors: improved beta environment, significant small-cap style rotation, and enhanced effectiveness of quantitative model factors [6] - The market environment remains resilient, but challenges related to style switching and volatility are anticipated [7] - The long-term advantages of quantitative index enhancement strategies lie in their risk control and portfolio optimization capabilities, allowing for dynamic adjustments to factor structures and risk exposures [7] Group 4: Future Market Outlook - The market is expected to transition from a phase of valuation recovery to one emphasizing profit realization, with increased uncertainty in style switching [12] - Despite potential challenges, the long-term configuration logic for quantitative index enhancement remains unchanged, supported by ample liquidity and policy backing [39] - The competitive landscape for quantitative private equity is likely to show a trend of polarization, with leading firms solidifying their positions while smaller firms seek differentiation [9][19]