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 ...

Core Viewpoint - Niche strategy funds are gaining recognition in the public fund industry, successfully breaking through traditional competition by exploring overlooked areas for excess returns [1][4][12]. Group 1: Performance of Niche Strategy Funds - The equity market has rebounded this year, leading to significant performance improvements for equity funds, particularly in mainstream sectors like technology and healthcare [3]. - Several niche strategy funds have achieved consistent positive returns over the years, with examples including 华夏新锦绣, 金元顺安元启, and 国金量化多策略, all maintaining positive returns for at least five consecutive years [4][5]. - 华夏新锦绣 fund, managed by 张城源, has achieved a 40.5% return this year and a cumulative return of 131.58% over five years [4]. - 金元顺安元启 fund, managed by 缪玮彬, has delivered a 29.41% return this year and a cumulative return of 262.3% over five years [5]. Group 2: Strategies Employed - Niche strategy funds utilize various strategies such as participating in private placements, quantitative stock selection, and tracking Smart Beta indices to uncover excess returns [4]. - The 定增 strategy, which involves participating in directed stock offerings at a discount, has shown promising results, with some stocks having over 50% floating profit for investors [4]. - Quantitative selection strategies have also been successful, as demonstrated by 国金量化多策略 fund, which achieved a 16.69% return this year [5]. Group 3: Market Dynamics and Company Growth - Smaller fund companies are leveraging their flexibility to quickly adapt and invest in niche strategies, leading to significant growth in fund sizes, such as 国金基金's equity fund size increasing from under 30 billion to nearly 130 billion [8]. - Larger fund companies like 华泰柏瑞 have also seen success with niche products, with their 红利低波ETF growing from 2.58 billion to 221.4 billion in size due to strong performance [9]. Group 4: Challenges Faced by Niche Strategy Funds - Niche strategy funds often face challenges such as "scale traps," where initial performance pressures can lead to significant fluctuations in fund size, risking liquidation [12]. - The effectiveness of niche strategies may require extended validation periods, and funds may be prematurely terminated during their development phase due to performance evaluations [13]. - Limited availability of niche strategy targets can lead to high concentration in holdings, increasing liquidity risks [14].

在公募基金行业，主流赛道向来是兵家必争之地，科技、大消费、医药等热门领域吸引了众多基金经理目光， 沉淀了大量资金。然而，一些基金凭借小众策略另辟蹊径，不追风口、不扎堆，总能在市场忽略的角落挖掘出 超额收益，成功破圈逆袭。 随着投资者认知提升，一些小众策略基金逐渐得到市场认可，近期规模一路攀升，不过也有部分基金因策略容 量限制而"长不大"。 小众策略破圈逆袭 一些小众策略基金凭绩优破圈。 今年以来，权益市场回暖，权益基金业绩大幅回升。科技、医药等一些主流热门行业获得基金经理大幅加仓， 并获得市场积极反馈，基金业绩也一路飙升。 与此同时，一些小众策略基金因为过去数年业绩积累，也逐渐获得投资者认可，成功破圈逆袭。 这些小众策略基金通过参与定增、量化选股、微盘选股、跟踪Smart Beta指数等策略，在市场忽略的角落挖掘 出超额收益，为投资者提供良好持有体验。 Wind数据显示，截至最新，仅有10余只权益基金连续5年保持正收益，其中不乏一些小众策略基金，如华夏新 锦绣、金元顺安元启、国金量化多策略、华泰柏瑞红利低波ETF等基金均实现了至少5年的连续正收益。 张城源管理的华夏新锦绣基金自2019年以来，每年均实现了正收 ...

Core Viewpoint - The public fund industry is witnessing a shift where niche strategy funds are gaining recognition and outperforming traditional funds by exploring overlooked market segments [1][2]. Group 1: Performance of Niche Strategy Funds - Niche strategy funds have seen significant performance improvements, with some achieving continuous positive returns over five years, such as 华夏新锦绣 and 金元顺安元启 [2][3]. - 华夏新锦绣 fund, managed by 张城源, achieved a 40.5% return this year and a cumulative return of 171.90% since 2020 [2]. - 金元顺安元启 fund, managed by 缪玮彬, has delivered a 29.41% return this year and a cumulative return of 389.56% since its inception in 2017 [3]. Group 2: Strategies Employed - Niche strategy funds utilize various strategies such as private placements, quantitative stock selection, and tracking Smart Beta indices to generate excess returns [2][3]. - The 定增 strategy, employed by 张城源, allows funds to acquire stocks at a discount through targeted placements, leading to significant gains post-lockup [2]. - Quantitative strategies, as seen in 国金量化多策略, have also shown consistent positive returns, with a 16.69% return this year [3]. Group 3: Market Dynamics and Company Strategies - Smaller fund companies are leveraging their flexibility to quickly adapt and invest in niche strategies, allowing them to capture market opportunities [4]. - 国金基金's assets grew from under 3 billion to nearly 13 billion due to its successful quantitative strategy [5]. - 华泰柏瑞基金's 红利低波ETF has seen its scale increase from 258 million to 22.14 billion, becoming the largest in its category due to strong performance [5]. Group 4: Challenges Faced by Niche Strategy Funds - Niche strategy funds often face challenges such as "scale traps," where initial performance pressures can lead to significant volatility and potential liquidation risks [7][8]. - The effectiveness of niche strategies may require extended validation periods, and funds may be prematurely terminated during their development phase [8]. - Some niche strategies are highly dependent on market conditions, making them vulnerable to changes in trends or policies [8].

Core Insights - The launch of the CSI 1000 index futures and options has marked a significant milestone in China's capital market reform, enhancing the product system for index futures and addressing the hedging needs for small-cap stocks [3][4] - The CSI 1000 index is expected to evolve into a "Chinese version" of the Russell 2000 index, focusing on small-cap growth stocks and providing diverse investment opportunities [7][9] Expansion of Index Funds - Since the launch of the CSI 1000 index futures and options, there has been a substantial increase in related index funds, with 41 out of 59 funds established post-launch, including 34 enhanced index products [6] - The CSI 1000 ETF's scale doubled within a year of the futures launch, with net inflows reaching 14.076 billion yuan [4] - The average daily trading volume of the CSI 1000 ETF surged by 597.5%, from 2.19 billion yuan to 15.28 billion yuan [4] Market Impact and Future Outlook - The CSI 1000 index is seen as a crucial tool for investors looking to tap into small and medium-sized enterprises, with a total market size of 148.6 billion yuan projected by mid-2025 [8] - The index's focus on small-cap stocks, particularly those classified as "specialized and innovative," is expected to attract more investment and enhance pricing efficiency [3][8] - The development of the CSI 1000 index ecosystem, including futures, options, and ETFs, is anticipated to provide more diverse investment options and improve market liquidity [9]

Core Viewpoint - The listing of the CSI 1000 index futures and options marks a significant milestone in the development of China's capital market, enhancing risk management tools and attracting more investments into small-cap stocks [3][4][10]. Group 1: Market Development - The CSI 1000 index futures and options have been instrumental in deepening capital market reforms, providing essential hedging tools for small-cap stocks and improving pricing efficiency [3][4]. - The introduction of these derivatives has led to a substantial increase in the scale and net inflow of related ETFs, with the South China CSI 1000 ETF's size doubling within a year of the futures listing, reaching a net inflow of 14.076 billion yuan [4]. - Daily trading volume of the South China CSI 1000 ETF surged by 597.5%, from 2.19 billion yuan to 15.28 billion yuan after the listing of the futures and options [4]. Group 2: Fund Expansion - Since the listing of the CSI 1000 index futures and options, there has been a steady expansion of related index funds, with 41 out of 59 funds established post-listing, including 34 enhanced index products [6][7]. - Over 80% of the 41 newly established CSI 1000 index funds have achieved positive returns, although there is significant size disparity among these funds, with 25 having assets below 200 million yuan [7]. - The CSI 1000 index funds are still in their early development stage, with limited recognition and small market capitalization of constituent stocks restricting fund sizes [7]. Group 3: Investment Outlook - The CSI 1000 index is viewed as a crucial tool for investors seeking opportunities in small-cap stocks, with a projected total ETF scale of 148.6 billion yuan by June 2025 [9]. - The index is expected to complement other major indices, providing a comprehensive coverage of A-share market styles and enhancing the investment ecosystem [9][10]. - The potential for the CSI 1000 index to become the "Chinese version" of the Russell 2000 index is supported by the anticipated continued release of profit elasticity in small-cap stocks amid economic recovery and industrial upgrades [10].

Core Viewpoint - The private equity stock strategy has achieved a return of 10% in the first half of the year, with quantitative strategies significantly outperforming subjective long strategies [1][2]. Summary by Sections Overall Performance - As of June 30, 2023, the average return of 10,041 private equity securities products was 8.32%, with over 80% achieving positive returns [3]. - Among various strategies, stock strategies led with an average return of 10%, followed by multi-asset strategies at 7.28%, and combination funds, bond strategies, and futures/derivatives strategies at 6.05%, 3.83%, and 3.82% respectively [3]. Quantitative vs. Subjective Strategies - Quantitative long strategies showed a remarkable return of 15.42%, while subjective long strategies had a return of 9.23% [9]. - In the first half of the year, 93.32% of quantitative long strategy products achieved positive returns, compared to less than 80% for subjective long strategies [9]. - The strong performance of quantitative strategies is attributed to their focus on small-cap stocks, which outperformed larger indices [9][10]. Market Conditions and Future Outlook - The market environment has been characterized by high trading volumes and volatility, benefiting quantitative strategies that capitalize on pricing discrepancies in small-cap stocks [9]. - Looking ahead, sectors such as military industry, artificial intelligence hardware and applications, and certain consumer electronics are expected to improve, presenting potential investment opportunities [7].

Core Viewpoint - The article highlights the growing significance of quantitative strategies in the investment landscape, particularly within private equity funds, showcasing their ability to generate excess returns compared to benchmark indices [2][3]. Group 1: Quantitative Strategies Overview - Quantitative strategies, especially quantitative long strategies, have become essential in the market, focusing on stock selection and optimization through models and algorithms to achieve excess returns [2]. - In May, 574 quantitative long products reported an average return of 3.77%, with an average excess return of 2.45%, indicating strong performance [2][3]. - The average excess returns for specific indices were as follows: CSI 300 at 0.97%, CSI 500 at 3.03%, and CSI 1000 at 2.84% [3]. Group 2: Performance of Specific Strategies - The top-performing products in the CSI 300 index over the past six months included those from Hainan Pengpai Private Equity and Ningbo Huansheng Quantitative, with excess returns of 6.81% and 5.67% respectively [4][5]. - For the CSI 500 index, the leading product was from Jintong Investment, achieving an excess return of 11.91% [8][10]. - In the CSI 1000 index, the top product was managed by Xiaoxiongmao Asset, with an excess return of 13.26% [10][12]. Group 3: Other Index Strategies - Other index products reported an average excess return of 14.41%, with the top performers coming from Liangchuang Investment and Longqi Technology [13][15]. - The strategy shift of certain products, such as the change from CSI 500 to other indices, has led to significant performance improvements [15]. Group 4: Quantitative Stock Selection - The average return for quantitative stock selection products was 9.83%, with an average excess return of 12.34% [17]. - The leading product in this category was managed by Zhuhai Zhengfeng Private Equity, achieving an excess return of ***% [19].

Quantitative Models and Factor Construction Quantitative Models and Construction Methods - **Model Name**: State-Owned Enterprise (SOE) Fundamental Factor Stock Selection Strategy **Model Construction Idea**: This model selects stocks from a pool of SOEs based on fundamental factors, emphasizing dividend characteristics and industry-specific metrics to evaluate profitability, operational efficiency, and solvency [3][5][6] **Model Construction Process**: 1. Define the SOE sample pool using the CSI SOE Index constituents and SOEs listed on the Beijing Stock Exchange for over six months [3] 2. Classify industries into two categories: dividend-oriented (e.g., resources, utilities, real estate) and growth-oriented (e.g., advanced manufacturing, software services) [3][4] 3. Select general factors such as ROE (TTM), operating cash ratio, asset-liability ratio, and dividend yield, alongside industry-specific factors like ROIC, inventory turnover, and R&D intensity [5][6][7] 4. Assign weights to factors, emphasizing dividend yield for all industries and adjusting weights for growth-oriented industries (e.g., lower weight for asset-liability ratio) [9] 5. Calculate scores for each stock based on weighted averages of general and industry-specific factors, normalize the scores, and rank stocks [10] 6. Allocate weights to the top 50 stocks using the formula: $$ w_{i} = \frac{score_{i}^3}{\sum_{i=1}^{N} score_{i}^3} $$ where \( score_{i} \) represents the normalized score of stock \( i \) [10] - **Model Evaluation**: The model effectively captures the dividend and growth characteristics of SOEs, providing a balanced approach to stock selection [3][5] - **Model Name**: Technology Theme Fundamental Factor Stock Selection Strategy **Model Construction Idea**: This model identifies technology stocks with high R&D intensity and strong fundamental performance, focusing on profitability, growth, and operational efficiency [17][18][21] **Model Construction Process**: 1. Define the technology stock pool based on industry classification (e.g., electronics, communication, computing) and R&D intensity (e.g., R&D expenses > 5% of revenue or R&D personnel > 10% of total employees) [17][18][19] 2. Exclude stocks in the "shakeout" and "decline" lifecycle stages, focusing on "introduction," "growth," and "maturity" stages [20][21] 3. Select fundamental factors, including general factors (e.g., gross margin growth, net profit growth) and unique factors (e.g., R&D expense ratio, PB-ROE) [22][23] 4. Calculate scores for each stock using the formula: $$ \hat{\mathbb{E}}_{i}^{s} = \frac{1}{5} Mean(S_{i}) + \frac{Mean(S_{i})}{Std(S_{i})} $$ where \( S_{i} \) represents the scores of eight factors for stock \( i \) [23][24] 5. Adjust scores using an R&D multiplier: $$ R&D \, Multiplier = 0.9 + 0.2 \times Normalization \left( \frac{Mean_{industry}(R&D/MarketCap)}{Mean_{A\_stock}(R&D/MarketCap)} \right) $$ Adjusted scores are then used to rank stocks [25][26] 6. Allocate weights to the top 50 stocks using the formula: $$ weight_{i} = \frac{score_{i}}{\sum_{i=1}^{50} score_{i}} $$ [27] - **Model Evaluation**: The model emphasizes R&D intensity and lifecycle stages, effectively identifying high-potential technology stocks [17][21] --- Model Backtest Results - **SOE Fundamental Factor Stock Selection Strategy**: - Annualized Return: 23.09% - Annualized Volatility: 21.77% - Sharpe Ratio: 1.0648 - Calmar Ratio: 0.9799 - Maximum Drawdown: -23.56% - Excess Return (vs. CSI SOE Index): 21.01% - Excess Sharpe Ratio: 1.7000 - Excess Calmar Ratio: 1.5867 - Excess Maximum Drawdown: -13.24% [11][12] - **Technology Theme Fundamental Factor Stock Selection Strategy**: - Annualized Return: 25.25% - Annualized Volatility: 28.22% - Sharpe Ratio: 0.9404 - Calmar Ratio: 0.7476 - Maximum Drawdown: -33.78% - Excess Return (vs. Technology Stock Pool): 10.62% - Excess Sharpe Ratio: 1.4755 - Excess Calmar Ratio: 1.2638 - Excess Maximum Drawdown: -8.40% [29][30] --- Quantitative Factors and Construction Methods - **Factor Name**: SOE General Factors **Factor Construction Idea**: Evaluate SOE performance using profitability, efficiency, and solvency metrics [5][6] **Factor Construction Process**: - Dividend Yield (TTM): Reflects SOE dividend stability - ROE (TTM): Measures profitability - Operating Cash Ratio: Indicates sales quality - Asset-Liability Ratio: Reflects financial stability - Labor Productivity: Measures operational efficiency [6] - **Factor Name**: Technology General and Unique Factors **Factor Construction Idea**: Assess technology stocks based on profitability, growth, R&D intensity, and supply chain metrics [22][23] **Factor Construction Process**: - Gross Margin Growth: Reflects profitability - Net Profit Growth: Indicates growth potential - R&D Expense Ratio: Measures R&D intensity - PB-ROE: Combines valuation and profitability - Supply Chain Metrics: Evaluate upstream and downstream risks [22][23] --- Factor Backtest Results - **SOE General Factors**: Incorporated into the SOE Fundamental Factor Stock Selection Strategy, contributing to its annualized return of 23.09% and Sharpe Ratio of 1.0648 [11][12] - **Technology General and Unique Factors**: Incorporated into the Technology Theme Fundamental Factor Stock Selection Strategy, contributing to its annualized return of 25.25% and Sharpe Ratio of 0.9404 [29][30]

- The short-term A-share models include the volume model (neutral), low volatility model (neutral), characteristic institutional model (neutral), characteristic volume model (bearish), intelligent CSI 300 model (bullish), and intelligent CSI 500 model (bearish) [1][10][68] - The mid-term A-share models include the limit-up and limit-down model (neutral) and the calendar effect model (neutral) [11][69] - The long-term A-share model is the long-term momentum model, which is neutral for all broad-based indices [12][70] - The comprehensive A-share models include the A-share comprehensive weapon V3 model (bearish) and the A-share comprehensive Guozheng 2000 model (bearish) [13][71] - The mid-term Hong Kong stock model is the turnover amplitude model, which is bullish [14][72] - The backtesting results for the models are as follows: - Volume model: neutral [1][10][68] - Low volatility model: neutral [1][10][68] - Characteristic institutional model: neutral [1][10][68] - Characteristic volume model: bearish [1][10][68] - Intelligent CSI 300 model: bullish [1][10][68] - Intelligent CSI 500 model: bearish [1][10][68] - Limit-up and limit-down model: neutral [11][69] - Calendar effect model: neutral [11][69] - Long-term momentum model: neutral [12][70] - A-share comprehensive weapon V3 model: bearish [13][71] - A-share comprehensive Guozheng 2000 model: bearish [13][71] - Turnover amplitude model: bullish [14][72]