Core Insights - The competition in quantitative investment is a long-term endeavor that requires continuous evolution to maintain a leading edge in a crowded market [1] - Starry Investment emphasizes a commitment to long-termism, leveraging technology for empowerment and maintaining a robust risk control framework [1][19] Company Overview - Founded in September 2020, Starry Investment quickly obtained a private fund management license and launched its first product, achieving over 10 billion in management scale by the end of 2021 [2] - The founder, Deng Jian, is a pioneer in applying artificial intelligence to quantitative strategy development, with a strong academic background and extensive industry experience [2][3] Investment Philosophy - The company's mission is to create a platform that emphasizes technological depth, rapid strategy iteration, and strict risk control, aiming to provide long-term value for investors [3] - The name "Starry Investment" is inspired by a classic poem, symbolizing the company's vision of exploring vast investment opportunities with a rigorous scientific approach [3] Research and Development Structure - Starry Investment has established a specialized research and development (R&D) team, with over 80% of its members holding doctoral degrees from top universities, ensuring a diverse academic background [6] - The company employs a unique investment manager (PM) responsibility system and a streamlined R&D process covering key quantitative research areas [4][5] AI Integration - The integration of AI technology is a core competitive advantage for Starry Investment, applied throughout the investment process, including factor mining and risk monitoring [7][8] - The company has developed a risk warning model that utilizes AI to predict short-term style returns, enhancing its risk management capabilities [7] Strategy Iteration and Optimization - Starry Investment has optimized its strategy iteration process, increasing the frequency of updates from quarterly to monthly, with core components being iterated every 2-3 weeks [10] - The firm employs a dual-track research model that combines deep learning with traditional multi-factor methods, enhancing the robustness and interpretability of its strategies [9] Risk Management - The company prioritizes compliance and risk control, establishing an automated and refined risk management system to identify unique market risks [12] - During market downturns, Starry Investment effectively managed excess drawdown risks, demonstrating the effectiveness of its risk control framework [12] Product Offering - Starry Investment has developed a comprehensive product line categorized into three main strategy types, catering to different risk-return preferences [13] - The focus on a proactive low-volatility product line distinguishes Starry Investment from competitors, aiming to provide stable long-term returns without frequent market timing [14] Future Outlook - The company aims to be a leader in long-term quantitative asset management in China, focusing on technology and risk control to create lasting value for investors [19] - Starry Investment is committed to continuous improvement in technology infrastructure and strategy development, adapting to industry changes and enhancing investor trust [19]

Core Insights - The "AI+Finance" Innovation Laboratory has been officially established in response to the national "Artificial Intelligence+" action plan, aiming to leverage the global fintech integration trend [1] - The first phase of the project, the "AI+Quantitative" Elite Training Camp, is set to launch, focusing on cultivating composite talents with international perspectives and solid theoretical foundations in AI and quantitative finance [1] - The training program is entirely free and offers necessary support for learning and living, along with substantial real-market funding for outstanding teams [1] Group 1 - The project adopts a dual-path training model of "quantitative strategy practice" and "fintech project development," allowing participants to choose based on their interests and expertise [2] - The training period lasts for 7 months, including 2 months of intensive coursework and project development, followed by 5 months of practical verification or project deepening [2] - The curriculum covers a comprehensive range of topics in quantitative investment, from financial data processing to machine learning applications in finance [2] Group 2 - The program aims to select 30 participants, primarily targeting doctoral and master's students, as well as senior undergraduates from universities in Beijing and Tianjin [2] - Applicants are required to have a background in interdisciplinary fields such as computer science, financial engineering, mathematics, statistics, physics, or electronic engineering, along with strong statistical and programming skills [2] - The official registration channel for the "AI+Quantitative" training camp and elite competition has opened, with a deadline set for January 15, 2026 [3]

二、分论坛：解码中国量化策略的下一增长曲线 在全球宏观格局深度演变、AI重构投研逻辑、新质生产力重塑产业格局的背景下，中国资管行业正站在"逻辑重塑"的十字路口。2026年1月15 日，由财视中国主办的第十九届HED中国峰会·深圳暨"第十九届介甫荣耀之夜"将于深圳盛大启幕，超过400位海内外金融机构掌舵人、顶尖投 资人与行业专家将齐聚一堂，共探2026年市场破局之道。 中国量化行业正站在一个鲜明的分水岭上，呈现出"量质齐升"与"竞争加剧"并行的新格局。规模上，百亿级量化私募数量已历史性地超越主观 多头同行；质量上，其超额收益能力日益稳健，2025年跑赢业绩基准的比例显著高于市场均值。但与此同时，赛道拥挤、AI技术竞赛加剧等挑 战也随之而来，出海寻求第二增长曲线已成为不少头部机构的共同选择——中资量化机构如何构建可持续的全球竞争力，成为行业亟待解答的 新课题。 针对这一背景，峰会特设分论坛"中国量化策略的下一阶段"，集结量化领域"梦之队"：千朔投资总经理丁志强、广金美好总经理罗山、大岩资 本总经理黄铂、优美利投资总经理贺金龙等重磅嘉宾，将围绕"量化股票策略的创新与前沿趋势""中资海外基金的生态构建"等核心议题深度 ...

Core Viewpoint - The article highlights the growing importance of quantitative fund managers in the financial market, emphasizing their reliance on mathematical models, algorithms, and big data analysis to create long-term value for investors. The demand for high-educated talent in this field has intensified due to advancements in AI, leading to a talent war among quantitative institutions [2]. Summary by Sections Education and Talent - Quantitative private equity funds favor highly educated professionals, with 69.11% of fund managers holding master's or doctoral degrees compared to 56.42% in subjective private equity [2]. Performance Overview - As of the end of November, there are 1,637 quantitative products with a total scale of approximately 135.11 billion, achieving an average return of 27.29% from January to November, significantly outperforming the market. Among these, 99 managers of billion-yuan private equity quantitative funds managed 386 products with an average return of 34.42%, yielding an excess return of 14.04% [3][4]. Performance by Fund Size - **100 Billion and Above**: 386 products with a total scale of 53.81 billion, average return of 34.42%, and 14.04% excess return [3]. - **50-100 Billion**: 165 products with a total scale of 17.49 billion, average return of 25.23%, and 10.31% excess return [8]. - **20-50 Billion**: 220 products with a total scale of 22.56 billion, average return of 26.62%, and 12.21% excess return [11]. - **10-20 Billion**: 176 products with a total scale of 12.60 billion, average return of 25.37%, and 10.10% excess return [14]. - **5-10 Billion**: 224 products with a total scale of 12.15 billion, average return of 25.75%, and 10.84% excess return [17]. - **0-5 Billion**: 466 products with a total scale of 16.52 billion, average return of 23.88%, and 11.19% excess return [19]. Top Performers - **100 Billion and Above**: Notable managers include Xu Jin and Lu Zhengzhe from Ningbo Huansheng, both achieving significant returns [4][5]. - **50-100 Billion**: Top managers include Shi En from Yunqi Quantitative and Huang Bo from Dayan Capital [8][10]. - **20-50 Billion**: Mo Bo from Luxiu Investment leads the performance [11][12]. - **10-20 Billion**: Wu Yintong from Longyin Tiger Roar is a top performer [14][15]. - **5-10 Billion**: Yan Xuejie from Huacheng Private Equity leads [17][18]. - **0-5 Billion**: Xie Libo from Jingying Zhito is at the forefront [19][20].

Core Viewpoint - The article discusses the pitfalls of the "Buy the Dip" strategy in investing, emphasizing that it often underperforms compared to a passive buy-and-hold approach and trend-following strategies [3][6]. Group 1: The Reality of Buying the Dip - The article highlights that over the past five years, investors have adopted a linear thinking approach: buying more as prices drop, believing that the market will eventually recover [3][4]. - AQR Capital Management's report analyzed 60 years of S&P 500 data and found that various dip-buying strategies underperformed compared to simply holding investments [10][11]. - The average Sharpe ratio for dip-buying strategies was lower than that of a buy-and-hold strategy, indicating a 16% reduction in risk-adjusted returns [11][12]. Group 2: Lack of Alpha in Dip-Buying - The report indicates that the average annualized alpha for dip-buying strategies was only 0.5%, with less than 8% of strategies showing statistically significant alpha [15]. - Holding investments for longer periods often leads to returns that reflect overall market performance rather than the effectiveness of the dip-buying strategy [19][20]. Group 3: The Flaws in Timing the Market - The article explains that dip-buying is essentially a value trade executed during a momentum phase, which often leads to poor timing and losses [21][26]. - Data shows a negative correlation between dip-buying strategies and trend-following strategies, suggesting that dip-buying often goes against market momentum [28][30]. Group 4: The Superiority of Trend Following - The article advocates for trend-following strategies, which have shown higher average annualized alpha compared to dip-buying strategies [31]. - During market downturns, trend-following strategies have historically provided better protection and even positive returns, contrasting sharply with the losses incurred by dip-buying strategies [35][36]. Group 5: The Ultimate Strategy: Portable Alpha - AQR proposes a "Portable Alpha" strategy that combines a long position in equities with a trend-following strategy, resulting in higher annualized excess returns and better risk-adjusted performance [41][42]. - This approach allows investors to benefit from market growth while also having a protective mechanism during downturns, effectively hedging risks [44][45]. Group 6: Practical Advice for Investors - The article concludes with three key recommendations for investors: avoid the temptation to time the market with dip-buying, respect market trends by incorporating trend-following strategies, and adopt a long-term investment perspective [49][54].

来源：一年打卡100场路演 诚奇量化总结：截至25年12月，规模470亿；两位管理人分别曾在千禧年和世坤工作；23年以后逐步转 向机器学习为主的非线性建模框架 贝小塔(VX:BetaRicher)整理, 仅供合格投资者审阅。 此举出于三方面考虑: 一是上海金融人才储备更丰富,便于招募高质量研究员; 二是员工更倾向在上海落户与发展,尤其在应届生招聘中更具吸引力; 三是上海作为国家金融中心,在政策支持与监管沟通上具备相对优势, 此次变更为非新设主体,统一社会信用代码及中基协登记编号均保 业务与投资运作不受任何影响。 截至25年12月初:基协显示该公司员工数为36人,其中 高度数量为3人。 正在运作产品为313个,延期清算为0,提前清算为104个,正常清算为6个。 公司曾于2022年底达到500亿规模高点,当前体置并未触及策略容量上限,在 现有约2万亿日均成交环境下运作舒适。内部设定若未来规模通近 700 亿将启动 封盘或参数调整机制,但目前多头策略额度充足,无主动控规模计划。 股权占比:股权渗透后分析得出,何文奇 占比 50.5089%,张万成 占比 49.4911%,两人合计绝对控股。为激励年轻化的核心投研队伍 ...

Core Insights - The conference highlighted the transformative role of AI in quantitative investment, emphasizing its potential to reshape research paradigms and enhance efficiency in the industry [1][2]. Group 1: AI's Impact on Quantitative Investment - AI's breakthrough, marked by Google's AlphaGo in 2016, has led to increased interest in applying AI technologies in investment, resulting in significant advancements [2]. - The reliance on experienced professionals in traditional quantitative investment has created high entry barriers, but AI and machine learning are reducing this dependency, thus redefining research paradigms [2]. - Despite the advantages, the application of AI is not infallible and requires human expertise for effective implementation [2]. Group 2: Practical Applications and Innovations - AI is becoming a focal point in quantitative trading, with companies like Square and Harmony utilizing deep learning models across various stages, from factor discovery to trade execution [3]. - The emphasis is on "incremental innovation" rather than "substitutive innovation," integrating AI into existing robust strategies while maintaining strict risk control [3]. - A closed-loop system combining model development, backtesting, risk control, and trade execution is essential for translating technological advancements into stable alpha [3]. Group 3: Challenges in AI Implementation - The quant market faces challenges such as strategy homogeneity, weak interpretability of AI models, and insufficient adaptability during extreme market conditions [4]. - The core issue lies in aligning the technical potential of AI with the fundamental nature of investment, which requires a balance between efficiency and risk control [4]. - The noise in financial data complicates predictions, indicating that neither AI nor human strategies are superior alone; instead, a collaborative approach is deemed the optimal resource allocation strategy [5].

Core Insights - The article emphasizes the importance of returning to the essence of finance and maintaining long-term competitive advantages in the increasingly competitive quantitative investment industry [1] - The firm "Shouzheng Yongqi" adopts a differentiated investment approach focusing on style timing as its core strategy, utilizing a three-dimensional framework of "style valuation - momentum - effective capital flow" to capture factor beta [1][2] Industry Landscape - The quantitative investment industry is experiencing a decline in entry barriers due to lower computing costs, widespread programming tools, and easier data access, leading to increased strategy homogeneity [1] - Current quantitative strategies are categorized into two types: popular multi-factor models that dominate the market and niche strategies based on professional financial understanding, which are more unique and capable of enduring through cycles [2] Competitive Barriers - The core competitive barrier for quantitative investment firms lies not in model tools but in the professional understanding of market styles, economic cycles, and capital behavior [2] - The proliferation of AI technology is expected to further differentiate these two models, with a significant portion of traditional quantitative fund managers potentially being replaced by AI, while those with deep professional insights will remain [2] Strategy Differentiation - "Shouzheng Yongqi" focuses on sustainable and stable positive returns, utilizing AI quantitative strategies developed from professional insights, contrasting with traditional multi-factor models that emphasize alpha (excess returns) [2][3] - The firm's unique style timing strategy emphasizes the importance of factor beta, assessing whether factors are bullish or bearish, and constructing a robust index enhancement system based on style trends [3] Risk Management - The firm's risk management capabilities are highlighted as a key indicator of model maturity, with the ability to identify risks in extreme market conditions and adjust factor exposures accordingly [3] - During liquidity crises, the firm's models successfully maintained lower drawdowns compared to similar models, demonstrating effective risk management [3] Market Outlook - The firm believes that the current market has significant upward potential and is in a rare phase of ample liquidity, presenting an optimal time for investment [3][4] - Investors are advised to focus on relative style valuations rather than chasing hot sectors, as overvalued sectors may present lower cost-effectiveness [4] - Within the technology sector, there are opportunities for rotation and switching between high and low valuations, with substantial growth potential in various sub-sectors [4]

Core Insights - The "2025 Quantitative Industry High-Quality Development Conference and Financial Technology·Quantitative Institution Golden Bull Award Ceremony" was held in Shanghai, recognizing Guangzhou Shouzheng Yongqi for its outstanding performance in the macro quantitative strategy category [1] - The Golden Bull Award is one of the most authoritative awards in China's capital market, aiming to select professional asset managers that can provide long-term stable returns to investors [1] - The Financial Technology Golden Bull Award focuses on recognizing institutions excelling in technology research and development, strategy iteration, risk control, and social responsibility within the financial technology and quantitative field [1] Group 1 - Dr. He Rongtian emphasized that large models do not inherently possess causal logic, stating that "correlation cannot predict the future; causality is the core of investment" [2] - He outlined a future direction for "AI + Quantitative" development, advocating for steady returns and innovative exploration rather than blindly pursuing technological singularities [2] - The investment philosophy in the AI era should focus on enhancing decision-making quality with AI technology while adhering to value investment principles [2] Group 2 - Dr. He expressed optimism about the A-share market, indicating that the current liquidity environment is the best in years and that there is still significant room for market development [2] - He highlighted the importance of relative valuation indicators and advised investors to avoid high-valuation stocks while considering long-term value investments [2] - In the technology sector, he noted that sub-sectors such as AI, new energy, and energy storage are experiencing rotation, with substantial growth potential in the long term [2]

Core Viewpoint - The role of traditional quantitative investment fund managers is expected to be replaced by artificial intelligence (AI), as stated by Feng Ji, founder of Beiyang Quantitative [1][2]. Group 1: AI's Impact on Quantitative Investment - AI is predicted to fundamentally transform quantitative investment, similar to how personal computers revolutionized the industry decades ago [2]. - Institutions that do not adopt AI technology may face elimination within the next 3 to 5 years [2]. - The advantages of AI in market learning and pattern recognition will become increasingly evident as computational power, data, and model capabilities improve [2]. Group 2: New Talent Requirements - The future of fund managers will require a new breed of professionals who are not only knowledgeable in investment research but also capable of translating these tasks into AI-related problems [2]. - The focus will shift from traditional investment research tasks to maintaining and developing advanced AI systems, necessitating a higher understanding of AI technology among professionals [2]. Group 3: Beiyang's Unique Approach - Beiyang Quantitative views the investment process as a unified machine learning task, contrasting with traditional methods that separate factors, signals, models, and strategies [3]. - The company employs a team primarily composed of engineers and computer scientists, lacking any employees with a financial background [3]. - Beiyang aims to replace expert experience with AI capabilities, asserting that AI can discover factors more effectively when provided with sufficient data [3]. Group 4: Competitive Advantages and Future Goals - Beiyang Quantitative claims three competitive advantages: high talent density in AI, superior computational power compared to any domestic university, and a self-developed AI experimental platform for real-time modeling and trading [3]. - The company aspires to evolve from a traditional quantitative private equity firm into an AI-native "computational company" [3]. - Beiyang's mid-term goal is to become a global quantitative manager, while the long-term vision is to establish itself as a significant computational entity [3].