最近公募基金圈子里发生了一件有趣的事，让我这个量化交易老手都忍不住要说道说道。2025年接近尾声，本该是基金经理们忙着 冲规模的时节，今年却玩起了新花样——一边是分红潮汹涌澎湃，一边是绩优基金纷纷限购。这葫芦里到底卖的什么药？ 作为一名量化交易者，我向来对市场异动特别敏感。Wind数据显示，截至12月4日，2025年以来3364只基金累计分红约2155.17亿 元。其中华泰柏瑞沪深300ETF以83.94亿元居首。与此同时，中欧旗下4只产品将单日申购限额降至1万元。这种看似矛盾的操作背 后，隐藏着机构资金的真实意图。 一、分红与限购：机构的两副面孔 记得我刚入行时，一位老前辈说过："市场就像个婊子养的赌场，但量化数据不会骗人。"这话虽然粗俗，但道理不假。现在的公募 基金就像个精明的老鸨，一边用分红吸引客户，一边又用限购把大客户拒之门外。 易方达科翔混合11月以来两次分红，合计1.05亿元。华夏基金说分红是盈利兑现，这话没错。但作为一个量化交易者，我更关心的 是：为什么现在分？为什么是这个金额？这些问题的答案都藏在数据里。 我翻看了近十年的基金分红数据，发现一个有趣的规律：大规模分红往往出现在市场转折点前。这不 ...

"赌徒式满仓" 暴露了对风险的漠视。小微账户中 90% 存在全仓押注行为，他们抱着 "一夜暴富" 的幻 想，将所有鸡蛋放在一个篮子里。广州股民小陈曾将 50 万本金全仓投入某题材股，起初赚了 20% 便自 诩 "股神"，但一次政策调整就让股价跌停，因无资金补仓只能被动躺平，最终亏损 40% 离场。上交所 数据显示，这类高频满仓交易者的亏损比例高达 93%，而高净值账户因分散配置，盈利比例达 60%。 "一买就跌，一卖就涨" 的魔咒，实则是散户被自身奇葩行为反噬的结果。从行为金融学视角看，四大 看似荒诞的操作背后，藏着认知偏差与情绪失控的深层密码，让真金白银沦为情绪的祭品。 "抄底逃顶强迫症" 源于对完美的执念。散户总想着买在最低点、卖在最高点，却忘了 "底是区域不是 点"。2025 年某科技股从 100 元跌至 60 元时，股民老王认定 "跌透了" 果断抄底，结果股价一路跌至 30 元，上演 "抄在半山腰" 的惨剧；而当这只股票反弹至 80 元时，他又因害怕回调匆忙卖出，错失后续翻 倍行情。这种总想赚尽最后一分钱的心态，反而让大部分收益擦肩而过。 "线性外推幻觉" 让散户沦为趋势的囚徒。股价大涨时认定 "会一 ...

一、数字狂欢背后的冷思考 最近科创板的两融数据又成了茶余饭后的谈资。2564.68亿元的总余额，东芯股份1.54亿元的净买入，这些数字在各大财经平台滚动播放。我盯着这些数 据看了许久，突然想起三年前那个燥热的夏天——当时创业板注册制刚落地，市场也是一片欢腾，但最终多少人真正从中分得一杯羹？ 数字会说话，但说的是加密语言。就像我清华实验室的同门师兄常说的："数据本身没有价值，解读数据的能力才值钱。"271只个股获得融资净买入，8 只超5000万元，这些数字背后藏着怎样的市场密码？普通投资者看到的可能是机会，而我看到的是一场认知维度的较量。 二、牛市幻觉与残酷现实 记得2015年那轮牛市，小区门口卖煎饼的大爷都开始讨论K线形态。当时有个现象特别有意思：80%的人确实赚过钱，但最终保住盈利的不足20%。这 就像在游乐场玩旋转木马——转得再欢，音乐停了才发现还在原地。 去年跟踪过一只半导体股票，走势堪称教科书级的"心理战"。股价在三个月里反复画"心电图"，论坛里骂声一片。但当我打开量化系统，看到的却是另 一番景象： 行情好的时候，多数人觉得"早涨晚涨都是涨"，结果往往是"赚过"而非"赚到" 机构用专业工具在收割认知差 ...

Group 1 - The article discusses the concept of "counterfactual thinking," which involves imagining alternative scenarios to past events and how this can lead to feelings of regret and self-blame [1][11] - It contrasts emotional counterfactual thinking, which focuses on a "better" past, with scientific counterfactual thinking, which aims to improve future decision-making by analyzing what could have been done differently [11][12] - The article emphasizes that scientific counterfactual thinking is essential for understanding causal relationships and making informed decisions, particularly in investment contexts [4][6][14] Group 2 - The article provides examples of how counterfactual thinking can be applied in various fields, such as vaccine safety, climate modeling, and engineering, highlighting its importance in evaluating potential outcomes and improving systems [2][9] - It explains that emotional counterfactual thinking often leads to negative feelings and a cycle of regret, while scientific counterfactual thinking encourages rational analysis and better decision-making [12][13] - The article concludes that mastering scientific counterfactual thinking can transform individuals from passive participants in their circumstances to active designers of their futures [15]

Quantitative Models and Construction Methods 1. **Model Name**: "Rediscover Confidence 2.0" RCP Factor - **Model Construction Idea**: The model is based on a common expectation bias in behavioral finance—overconfidence. It innovatively uses high-frequency minute sequence data to construct the overconfidence factor CP by calculating the time gap between favorable price surges and price corrections. The second-generation Rediscover Confidence Factor (RCP) is derived by orthogonalizing the first-generation CP factor with intraday returns and using the residuals as the RCP factor[1][6]. - **Model Construction Process**: - **Step 1**: Calculate the time gap between favorable price surges and price corrections to construct the overconfidence factor CP. - **Step 2**: Orthogonalize the CP factor with intraday returns. - **Step 3**: Use the residuals from the orthogonalization process as the second-generation Rediscover Confidence Factor (RCP)[6]. - **Model Evaluation**: The RCP factor constructed based on the Rediscover Confidence idea performs significantly better than traditional combination methods[6]. Model Backtesting Results 1. **"Rediscover Confidence 2.0" RCP Factor**: - Annualized Return: 17.68%[1][7][12] - Annualized Volatility: 7.83%[1][7][12] - Information Ratio (IR): 2.26[1][7][12] - Monthly Win Rate: 77.46%[1][7][12] - Maximum Drawdown: 7.46%[1][7][12] Quantitative Factors and Construction Methods 1. **Factor Name**: Overconfidence CP Factor - **Factor Construction Idea**: The factor is based on the degree of investor overconfidence affecting stock prices, using the time difference between rapid price increases and decreases as a proxy variable[6]. - **Factor Construction Process**: - **Step 1**: Calculate the time difference between rapid price increases and decreases to construct the overconfidence factor CP[6]. - **Factor Evaluation**: The CP factor innovatively captures the overconfidence bias in investor behavior[6]. 2. **Factor Name**: Rediscover Confidence RCP Factor - **Factor Construction Idea**: Considering that investors may become overly pessimistic during price corrections, leading to excessive corrections, but due to favorable news, such stocks will eventually rebound. The RCP factor is derived by orthogonalizing the CP factor with intraday returns and using the residuals[6]. - **Factor Construction Process**: - **Step 1**: Orthogonalize the CP factor with intraday returns. - **Step 2**: Use the residuals from the orthogonalization process as the Rediscover Confidence Factor (RCP)[6]. - **Factor Evaluation**: The RCP factor, after purification, shows significantly improved performance[7]. Factor Backtesting Results 1. **Rediscover Confidence RCP Factor**: - IC Mean: 0.04[1] - Annualized ICIR: 3.27[1] - Annualized Return: 20.69%[1] - Information Ratio (IR): 2.91[1] - Monthly Win Rate: 81.55%[1]

一、健康科技独角兽的资本盛宴 11月28日那天，我正盯着量化系统里跳动的数据流，突然弹窗跳出轻松健康通过港交所聆讯的消息。这家注册用户1.684亿的数字健康巨头，用54.9% 的年复合增长率向市场展示着什么叫"科技+保险"的黄金赛道。作为见证过无数IPO的量化交易者，我却注意到一个有趣现象：每当这类明星企业上 市，普通投资者的账户往往与市场热度形成诡异反差。 记得2025年上半年，某医疗AI企业上市首日暴涨47%，但同期调查显示78%的散户投资者在这波行情中收益率不足5%。这种割裂感让我想起自己十 年前初入市场时的困惑——明明利好消息铺天盖地，为什么实际操作起来总是差之毫厘？ 二、4000点牛市里的冰冷数据 上证指数突破4000点那天，我的朋友圈被各种庆祝刷屏。但量化系统给出的数据却呈现另一个故事：4月7日至10月30日期间，虽然指数上涨19.6%， 但仅有40%个股跑赢大盘。更值得玩味的是，4200只上涨个股中超过4000只振幅大于30%，这意味着大多数投资者其实是在过山车行情中反复被收 割。 这让我想起行为金融学中的"处置效应"——投资者总是过早卖出盈利股票而过久持有亏损股票。在量化视角下，这种现象的本质 ...

Core Insights - The primary challenge for the expansion of the AI industry in the U.S. is the shortage of electricity, with a projected demand of 69 GW by 2028 and a shortfall of 44 GW, equivalent to 44 nuclear power plants [1][2] - The construction cost for each additional 1 GW of data center capacity is approximately $50 billion, leading to concerns about whether the industry is entering an investment bubble [1][2] - The discussion revolves around two main questions: where will the electricity come from, and how will the funding for this massive infrastructure be secured [1][2] Electricity Shortage Solutions - The first conventional method to address the electricity shortage is the transition of cryptocurrency miners to AI data centers, which could potentially release 15 GW of power within 18-24 months [1][2][6] - Nuclear power is considered a long-term solution, with conventional plants taking over ten years to build, while small modular reactors (SMRs) are not expected to be commercially viable before 2030-2035 [2][3] - Natural gas is another option, but the supply of gas turbines is limited, with a backlog of 2-4 years for orders, making it a challenging short-term solution [4][5] - Fuel cell storage and solar plus storage are also mentioned, but they are not expected to provide immediate relief [5][6] Financing the AI Infrastructure - The financing landscape is complex, with companies like CoreWeave facing significant debt and high-interest rates, indicating a reliance on external funding [16][18] - Investment-grade bonds are expected to be a primary source of financing, with estimates suggesting that the high-rated market could address $300 billion in funding needs next year and $1.5 trillion over five years [26][28] - Asset-backed securities (ABS) and collateralized debt obligations (CDOs) are potential financial instruments that could be utilized to package and sell the underlying assets of data centers [19][20] Market Dynamics and Competition - NVIDIA is positioned as a central player in the GPU market, with its products being critical for AI data centers, while AMD is seen as a competitor trying to gain market share [30][31] - OpenAI is viewed as a disruptive force, driving demand for GPUs and influencing the strategies of other major tech companies [31][32] - The behavior of large tech companies is influenced by the fear of missing out on potential breakthroughs in AI, leading to significant investments despite the risks [33][34] Transition of Cryptocurrency Miners - The transition of cryptocurrency miners to AI data centers is seen as a viable solution, with early movers like CoreWeave benefiting from their timely shift [40] - New entrants in the market may face challenges due to their previous reliance on Bitcoin mining, which could complicate their transition to AI data centers [40]

Core Viewpoint - The Federal Reserve is experiencing internal divisions regarding monetary policy, with a focus on balancing inflation control and employment stability, likened to bargaining in a market [1][2]. Group 1: Federal Reserve's Position - San Francisco Fed President Mary Daly highlighted the risk of "non-linear" deterioration in the job market, suggesting that current stability could quickly change [2]. - Daly described the current state as a "low hiring, low firing" balance, which is precarious and requires preventive measures despite inflation not being fully controlled [2][12]. - The ongoing debate within the Fed about interest rate cuts reflects a deeper struggle among various economic forces [7]. Group 2: Market Dynamics - The article emphasizes the importance of quantitative tools for retail investors to understand market dynamics, similar to how Daly uses data to assess economic trends [2][12]. - Historical examples of stocks that experienced prolonged consolidation before significant price movements illustrate the hidden market dynamics at play [5][12]. - The presence of "hot money" signals in stock movements can indicate potential market shifts, suggesting that investors should pay attention to underlying capital flows rather than surface price changes [12][13]. Group 3: Behavioral Finance Insights - Daly's comments on groupthink highlight the risks of consensus in market sentiment, where widespread bullishness may signal caution [8][11]. - The concept of "anchoring effect" in behavioral finance suggests that investors often misinterpret price fluctuations, overlooking the underlying capital movements [11]. Group 4: Recommendations for Investors - Investors are advised to utilize tools and maintain patience in the face of uncertainty, focusing on current risks rather than future uncertainties [13]. - Monitoring the activity of both retail and institutional investors can provide insights into potential market movements, indicating when to start paying attention to emerging trends [13]. Group 5: Future Outlook - Anticipation of increased market volatility as the December FOMC meeting approaches, with a clear message that only those equipped with advanced tools will navigate this environment effectively [14][15]. - Understanding the real movements of capital behind stock price fluctuations is crucial for making informed investment decisions [15].

二、牛市暴跌的悖论 这些年我最大的感悟就是：牛市多暴跌。2019年创业板那根近8%的大阴线，2020年三根阴线就跌去10%的惨烈场景仍历历在目。这种看似反常的现象背 后，实则暗藏玄机。 人天生厌恶损失的心理特质，使得我们在面对剧烈调整时容易惊慌失措。但有趣的是，这些"牛市的暴跌"往往成为主力资金最好的掩护——既可以是资金出 清的烟幕弹，也可能是洗盘震仓的利器。就像海菲曼那看似完美的财务数据下，谁又能说清机构资金的真实意图？ 三、两个典型案例的启示（一）温水煮青蛙式的折磨 一、北交所新贵引发的思考 昆山海菲曼科技即将登陆北交所的消息，让我这个浸-市场多年的量化投资者不禁莞尔。这家拟募资4.3亿元的高端电声技术企业，其发展轨迹恰似A股市场 那些令人又爱又恨的牛股——表面光鲜亮丽，内里暗流涌动。 看着海菲曼前三季度29.49%的净利润增速和216项专利的技术壁垒，我不禁想起这些年见证过的无数"科技新贵"。它们往往带着耀眼的光环上市，却在随后 的市场波动中让无数散户尝尽酸甜苦辣。这让我意识到：与其追逐热点IPO，不如静下心来研究市场波动的本质规律。 橙色柱体显示的"机构库存"数据始终活跃，证明所有震荡都是主力精心设计 ...

Market Overview - A-shares exhibited a typical divergent trend today, with the China Shipbuilding System concept experiencing significant gains, particularly China Shipbuilding Defense hitting the daily limit, while the commercial aerospace concept also saw a surge [3][6] - The new stock Moer Thread, focused on GPU development, attracted considerable attention during its subscription [3] Financing Data - Despite an overall decline in financing balance for three consecutive days, 25 stocks received net financing inflows exceeding 50 million yuan, with Dekeli leading at 156 million yuan [3][4] - Other notable stocks with significant net financing include Beijing Bank at 151 million yuan and Zhongwen Online at 141 million yuan [4] Behavioral Finance Insights - The phenomenon of more severe adjustments in bull markets compared to bear markets is attributed to loss aversion, where investors experience greater pain from losses than pleasure from equivalent gains [5] - Institutional funds exploit this psychological weakness, using volatility to disrupt retail investors' resolve [6] Investment Strategy - The current market conditions reflect strategic repositioning by institutional funds, with stocks receiving large financing inflows likely to be long-term targets for these institutions [13] - The analysis emphasizes the importance of recognizing the nature of trading behaviors, distinguishing between genuine institutional activity and retail speculation [9][14] Sector Performance - The military and aerospace sectors have seen continuous institutional support for three months, indicating a strong interest in these areas [14] - The GPU and other hard technology sectors are identified as long-term strategic tracks, suggesting potential for future growth [14]