Core Viewpoint - The recent increase in financing funds, reaching approximately 2.6 trillion yuan, indicates a positive market sentiment, but surface price movements may not reflect true trading intentions [1] Group 1: Market Behavior - Many investors are misled by superficial price fluctuations, often reacting emotionally to market movements, which can lead to poor decision-making [3][9] - There are two types of market behaviors: one where prices rise but experience multiple pullbacks, leading investors to sell out of fear; and another where prices appear to rebound, misleading investors into thinking new funds are entering the market [5][8] Group 2: Quantitative Data Analysis - Utilizing quantitative data can help reveal the true trading intentions behind market movements, as traditional observations may not capture the underlying dynamics [7][8] - The "institutional inventory" data reflects the activity level of large institutional funds, providing insights into whether institutions are actively participating in trades [6][8] Group 3: Investment Strategy - A continuous increase in financing funds suggests a generally positive market attitude, but it is crucial to analyze individual stocks beyond their price movements, focusing on the underlying trading behaviors [8] - If a stock shows price fluctuations but maintains active "institutional inventory," it indicates significant institutional involvement; conversely, if a stock rebounds without such activity, it may be a false signal [8][9]

1. Report Industry Investment Rating - Not provided in the content 2. Core Viewpoints of the Report - The scale of index - enhancing products significantly increased in 2025, with A500 and non - traditional broad - based index - enhancing products growing notably. The excess returns of index - enhancing products fluctuated, and the differentiation degree within each broad - based index - enhancing category became larger [7][24]. - The scale of active quantitative funds also grew, with the full - industry quantitative stock - picking strategy and the quantitative products of the active equity team showing obvious scale growth [40]. - The market actively embraced quantitative fixed - income + funds in 2025, with the strategy pool becoming more diverse [76]. 3. Summary According to the Directory 3.1 Index - enhancing Funds - **Scale & New Issuance**: By 25Q3, the scale of index - enhancing products exceeded 2500 billion yuan, with A500 and non - traditional broad - based index - enhancing products having significant scale growth. In 2025, some non - conventional broad - based index - enhancing products in new issuances had large scale. The top 10 custodian banks and fund companies with large new - issuance scales were identified [7][10]. - **Fund Company Statistics**: As of 25Q3, 23 fund companies had index - enhancing product management scales of over 30 billion yuan. E Fund had the largest management scale. Different fund companies had different advantages in various types of index - enhancing products [13]. - **Excess Return Performance**: The Alpha effect of index - enhancing products peaked in 2020 and then declined. In 2025, the excess returns of three major types of index - enhancing products fluctuated, and the differentiation degree within each type increased. Small - cap index - enhancing products had higher excess returns [21][24]. - **High - performing Index - enhancing Products**: Some high - performing index - enhancing products had excess returns exceeding 20% in 2025. Many high - performing products had good performance adaptability in various market environments and had specific factor exposure characteristics [30][32]. - **Index - enhancing Product Watchlist**: The report selected fund products that were superior in their respective types based on multi - dimensional investment ability evaluations [36]. 3.2 Active Quantitative Funds - **Seven Strategy Types & Scale Changes**: Active quantitative funds can be divided into seven major categories. The full - industry quantitative stock - picking strategy and the quantitative products of the active equity team had obvious scale growth in 2025 [39][40]. - **Similar Index - enhancing Strategy**: The partial - equity fund index - enhancing strategy received high attention. Different products had different strategies when benchmarking the partial - equity fund index [46]. - **SmartBeta Strategy - Micro - and Small - cap**: The micro - and small - cap strategy can be classified into three major types. The difference in the degree of market - value sinking of stocks led to different returns [50][51]. - **SmartBeta Strategy - Dividend**: In 2024 and 2025, many public funds deployed dividend - strategy products. Companies sought differentiated layouts, and the investment in Hong Kong stocks had a significant impact on the performance in 2025 [53]. - **SmartBeta Strategy - Growth**: Different growth - style funds adopted different investment strategies, with different industry preferences and market - value exposures [56]. - **SmartBeta Strategy - Value**: Different value - style funds adopted different investment strategies, focusing on different aspects such as value - growth attributes and multi - strategy investment [61]. - **Full - industry Stock - picking Strategy**: The full - industry quantitative stock - picking strategy was diverse, including industry rotation, factor rotation, and multi - strategy [63]. - **Integration of Active and Quantitative**: Some fund managers actively integrated active and quantitative strategies, with different product strategies and positioning [69]. 3.3 Quantitative Fixed - income + Funds - **Scale & New Issuance**: There were about 171 quantitative fixed - income + funds in the market in 2025, with a scale increase of 36.7 billion yuan and a total scale of about 122.547 billion yuan. The market actively embraced quantitative fixed - income + strategies [76]. - **Index - enhancing Strategy**: The index - enhancing strategy in fixed - income + funds provided a tool - type product for obtaining broad - based index beta returns. The effectiveness of the strategy was related to multiple factors, and some fund companies had a large layout in this area [84]. - **Style Strategy**: The style strategy evolved from the value style to the growth style and the barbell strategy. Some companies innovated in this area to meet different market demands [88]. - **Convertible Bond Quantitative Strategy**: The convertible bond quantitative strategy was represented by E Fund's Dual - Bond Enhancement, which used a convertible bond option - pricing model for statistical arbitrage [90]. - **Market Dynamics**: Many active - management fixed - income + fund managers actively embraced quantitative investment, such as China Europe Fund, E Fund, and GF Fund [92]. - **Quantitative Fixed - income + Fund Watchlist**: The report screened out quantitative fixed - income + funds at different volatility levels based on multiple indicators [107].

Report Overview - The report is titled "Public Offering Quantitative Funds: 2025 Annual Strategy Review and 2026 Annual Strategy Outlook" and was released on January 12, 2026 [1] Industry Investment Rating - No industry investment rating information is provided in the report Core Views - In 2025, the scale of index - enhanced products significantly increased, with the total scale reaching 257.2 billion yuan by Q3 2025. The excess returns of index - enhanced products fluctuated, and the differentiation within each broad - based index enhancement became more obvious. Some high - performing index - enhanced products showed good adaptability to various market environments and had specific factor exposures [2][23] - Active quantitative funds can be divided into seven major categories, and the scale of all - industry quantitative stock - selection strategies and active equity team quantitative products increased significantly in 2025 [2][42] - The market actively embraced quantitative fixed - income + funds in 2025, with the total scale increasing by 36.7 billion yuan to about 122.547 billion yuan. The strategy pool of these funds became more diverse, including index - enhancement, style, convertible bond quantitative, and active - quantitative combination strategies [2][83] Section - by - Section Summaries 1. Index - Enhanced Funds 1.1 Scale & New Issuance - By Q3 2025, the scale of index - enhanced products exceeded 250 billion yuan, reaching 257.2 billion yuan, a significant increase from 206.5 billion yuan in Q4 2024. Non - traditional broad - based index - enhanced products such as A500 index - enhanced and some non - traditional broad - based index - enhanced products (e.g., ChiNext Index Enhancement) had significant scale growth [8] - In 2025, some non - conventional broad - based index - enhanced products had large new - issuance scales, such as GF ChiNext Index Enhancement with a new - issuance scale of 2.393 billion yuan [8] 1.2 Fund Company Statistics - As of Q3 2025, E Fund had the largest management scale of index - enhanced products, totaling 26.733 billion yuan. Companies with relatively comprehensive index - enhanced product lines included Huatai - PineBridge and Fullgoal [15] - In different types of index - enhanced products, some fund companies had outstanding performance in 2025. For example, Southern Fund's products in the SSE 50 index - enhanced category had an average excess return of 10.15% [18] 1.3 Excess Return Performance - The Alpha effect of index - enhanced products peaked in 2020 and then declined. In 2025, the excess returns of three major types of index - enhanced products (CSI 300, CSI 500, and CSI 1000) fluctuated significantly. The excess return of CSI 1000 index - enhanced products was strong in the first half of 2025, while those of CSI 300 and CSI 500 were weak. By the fourth quarter, most excess returns recovered [23] - The differentiation within each broad - based index enhancement was greater in 2025, especially in the CSI 1000 index - enhanced products, where the standard deviation of excess returns exceeded 6%, and the performance difference between the best - and worst - performing products was close to 25% [25] 1.4 High - Performing Index - Enhanced Products - In 2025, some high - performing index - enhanced products had significant excess returns. For example, Furong CSI 300 Enhancement had the best performance among CSI 300 index - enhanced products, and ICBC Credit Suisse CSI 1000 Index Enhancement performed well among CSI 1000 index - enhanced products [33] - Many high - performing index - enhanced products showed good performance in various market environments. Commonly positively exposed factors included growth, dividend, profitability, and analyst factors, while negatively exposed factors included volatility, liquidity, market capitalization, and valuation [35][37] 1.5 Index - Enhanced Product Watch List - The report selected fund products that were dominant in their respective types based on multi - dimensional investment ability evaluations, considering factors such as the ability to convert trading turnover into returns, the stability of Alpha acquisition, and performance stability in various market environments [39] 2. Active Quantitative Funds 2.1 Seven Strategy Types & Scale Changes - Active quantitative funds can be divided into seven major categories: all - industry quantitative stock - selection, active equity team quantitative, style funds, quasi - index - enhanced funds, industry - themed funds, industry - rotation funds, and Hong Kong stock quantitative funds [42] - The scale of all - industry quantitative stock - selection strategies increased significantly, followed by active equity team quantitative products. The total scale of active quantitative funds in Q3 2025 was 227.895 billion yuan [43] 2.2 Quasi - Index - Enhanced Strategy - Some quasi - index - enhanced strategies targeted the CSI 300, CSI 500, or the active equity fund index. Products such as Bodaoyuanhang and Bodaojiuhang targeted the active equity fund index, but they had different strategies [49] 2.3 SmartBeta Strategy: Small - and Micro - Cap - The small - and micro - cap SmartBeta strategy can be divided into three categories: more focused on micro - cap stocks, more focused on small - cap stocks, and similar to CSI 2000 index - enhancement strategies. The degree of market - capitalization decline of stocks affected the product returns [53] 2.4 SmartBeta Strategy: Dividend - In 2024 and 2025, many public - offering funds launched dividend - strategy products. Some companies sought differentiated layouts, such as Ruidaxinhong Quantitative 6 - Month Holding with a market - capitalization decline in dividend stocks and GF High - Dividend Preference focusing on specific company screening and Hong Kong stock dividend investment opportunities [56] 2.5 SmartBeta Strategy: Growth - Different growth - style active quantitative funds had different investment strategies. For example, Bodaogrowth Zhihang used a multi - factor stock - selection enhancement model based on the CITIC Growth Style Index, and GF New - Generation Selection focused on selecting high - growth stocks [59][60] 2.6 SmartBeta Strategy: Value - Value - style active quantitative funds, such as GF Value Pilot One - Year Holding, combined subjective fundamental research and stock - selection with a value - growth style. Other products, like Caitong Huazhen Quantitative Stock - Selection, targeted specific benchmark indices [65] 2.7 All - Industry Stock - Selection Strategy - The all - industry quantitative stock - selection strategy was diverse. Products such as Guojin Quantitative Multi - Factor used factor - rotation strategies, China Merchants Quantitative Selection used a PB - ROE framework, and Hua'an Event - Driven Quantitative Strategy adopted an industry - rotation strategy [67][69] 2.8 Integration of Active and Quantitative - Some fund managers, such as Yang Dong of GF Fund and Zhang Xueming of China Europe Fund, integrated active and quantitative strategies in their product management. Their products had different strategy positioning and characteristics [74][77] 3. Quantitative Fixed - Income + Funds 3.1 Scale & New Issuance - There were about 171 quantitative fixed - income + funds in the market in 2025, with the total scale increasing by 36.7 billion yuan to about 122.547 billion yuan. The top - ranked funds in terms of scale reached tens of billions of yuan [83] - In 2025, fund companies paid high attention to quantitative fixed - income + funds, and about 1/5 of the new - issued products belonged to quantitative strategies [88] 3.2 Index - Enhancement Strategy - Fixed - income + funds using index - enhancement strategies in the equity part provided beta returns of broad - based indices. The effectiveness of the strategy was related to the characteristics of the benchmark index, investment value, product type, and position - central setting [92] 3.3 Style Strategy - The style strategy of fixed - income + funds evolved from value - style to growth - style and barbell strategies. Some products, such as China Europe Dingli, adopted a boom - growth strategy, while others used barbell strategies that combined dividends and growth [96] 3.4 Convertible Bond Quantitative Strategy - E Fund Dual - Bond Enhancement was a representative product of convertible bond quantitative strategies, using a convertible bond option - pricing model for statistical arbitrage [98] 3.5 Market Trends - Many fund companies' active - management fixed - income + fund managers actively embraced quantitative investment. For example, China Europe Fund promoted the "industrialization" of the investment - research system and developed a four - factor SmartBeta strategy, and E Fund's Bao Zhengyu combined active research and quantitative models [100][111] 3.6 Quantitative Fixed - Income + Fund Watch List - The report selected quantitative fixed - income + funds with different volatility levels as the watch list based on factors such as risk - control ability, return stability, and performance sustainability [115]

Core Insights - The "Entropy Machine" AI model, the first financial meteorological AI model in China, was launched at the second Financial Meteorology Academic Annual Conference in Guangzhou, developed by Fudan University and the National Meteorological Information Center [1][3]. Group 1: Model Overview - The "Entropy Machine" model is constructed based on global meteorological reanalysis data and stock price-volume data, capable of predicting short-term returns for the majority of A-share market stocks [3]. - Validation results indicate that the model effectively identifies industries highly sensitive to meteorological factors, such as renewable energy (wind and solar power), traditional petroleum and chemical industries, construction, and agriculture, aligning with classifications from the World Meteorological Risk Management Association [3]. Group 2: Investment Strategy and Applications - Investment strategies developed from the model's testing and inference results have shown consistent and stable positive returns during historical backtesting across multiple time periods [3]. - The model has broad applications in the financial sector, allowing listed companies in meteorologically sensitive industries to manage climate risks and maintain market value; banks and insurance companies can utilize it for risk control in equity pledge businesses and explore innovative climate financing models; investors can use it as a quantitative investment tool; and academia can leverage model outputs to test and refine asset pricing theories [3]. Group 3: Purpose and Future Implications - The release of the model aims to explore the role of meteorological factors in financial asset pricing, providing innovative tools for risk management and investment decision-making [3]. - Its application is expected to support the construction of intelligent financial service systems and quantitative assessments of meteorological risks [3].

Core Insights - The Shanghai Composite Index has shown a strong upward trend since mid-December 2025, achieving a record 16 consecutive days of gains by January 9, 2026 [1][15] - In December 2025, the Shanghai Composite Index, Shenzhen Component Index, and ChiNext Index increased by 2.06%, 4.17%, and 4.93% respectively, although none reached new yearly highs [15] - Commodity futures such as gold, silver, copper, aluminum, and lithium carbonate experienced significant price increases in December 2025 [15] - A total of 2,362 private equity products from 950 firms reached historical net value highs in December 2025, with 132 firms achieving this across all their products [15] Private Equity Performance - Among the 132 firms with all products reaching historical highs, 96 firms had at least three products displayed on the private equity performance platform for 2025 [15][16] - The breakdown of the 96 firms by core strategy shows 46 focused on stock strategies, 25 on futures and derivatives, and 15 on multi-asset strategies [2][15] - The investment modes of these firms include 42 subjective, 34 quantitative, and 20 mixed strategies [3][15] Billion-Level Private Equity - 12 private equity firms with over 10 billion yuan in assets achieved historical highs in December 2025, with eight being quantitative firms and eight focusing on stock strategies [4][16] - The top firms by average returns for 2025 include Lingjun Investment, Ningbo Huanfang Quantitative, and Chengqi Private Equity [4][19] 20-100 Billion Private Equity - 15 private equity firms in the 20-100 billion yuan range also reached historical highs, with eight being subjective and six quantitative [7][20] - Leading firms by average returns include Zhihua Asset Management, Hengbang Zhaofeng, and Yunqi Quantitative [7][21] 5-20 Billion Private Equity - 31 private equity firms in the 5-20 billion yuan range achieved historical highs, with 15 being subjective and 10 quantitative [9][23] - Top firms by average returns include Shanghai Hengsui Asset, Sapphire Fund, and Huacheng Private Equity [9][24] Below 5 Billion Private Equity - 38 private equity firms with assets below 5 billion yuan reached historical highs, with 18 being subjective and 10 quantitative [12][26] - Leading firms by average returns include Jingying Zhito, Fanxu Asset, and Sanhua Asset [12][27]

- The report does not contain any specific quantitative models or factors for analysis [1][2][3] - The report primarily focuses on market trends, valuation metrics, and style performance without detailing quantitative models or factor construction [1][2][3] - No formulas, construction processes, or backtesting results for quantitative models or factors are provided in the report [1][2][3]

Summary of Key Points Core Viewpoints - The fourth quarter of 2025 saw mixed performance across different indices, with the Growth Index rising by 5.03% and the Shanghai Composite Index increasing by 2.22% [1] - Most sectors within the CITIC first-level industries showed positive returns, particularly the Oil & Petrochemical sector, which rose by 16.97%, and the Defense & Military sector, which increased by 16.74% [1][10] - Timing models in the fourth quarter demonstrated the ability to achieve absolute positive returns, with several models performing notably well [1] Sector Performance - The Oil & Petrochemical sector had a closing price of 3,424.25 with a quarterly increase of 16.97% [11] - The Defense & Military sector closed at 11,864.34, reflecting a quarterly rise of 16.74% [11] - Other sectors with significant gains included Nonferrous Metals (15.63%), Communications (14.72%), and Consumer Services (8.45%) [11][10] Fund Performance - Balanced mixed funds outperformed others with an average return of 1.22%, while stock funds showed a decline of 1.71% [14] - A total of 715 new public funds were established in Q4 2025, raising a total of 2,784.53 billion, with mixed funds raising 997.56 billion [14] Investment Themes - The report emphasizes the importance of utilizing historical timing, sector rotation, and stock selection models to identify future investment opportunities [5][6] - The focus for Q1 2026 is on sectors such as Construction Materials, Automotive, and Electronics [3] Timing Strategies - The report outlines various timing models, including short-term models like the Price-Volume Resonance Model and the Low-Volatility Blade Model, which aim to capture market trends and rebounds [15][16] - The Long-term Momentum Swing Model has shown a 7.01% annualized return since June 2008, indicating its effectiveness in long-term market analysis [43][44] Comprehensive Models - The Comprehensive Weapon V3 Model integrates multiple timing strategies and has achieved an annualized return of 29.55% since February 2015 [46] - The Smart Algorithm Timing Model for the CSI 300 Index has demonstrated a remarkable annualized return of 35.42% since January 2014, showcasing its superior performance compared to the index itself [49][50]

Core Viewpoint - In 2025, the Chinese market showed significant growth, with major stock markets rising sharply and structural opportunities in various sectors, particularly in technology growth stocks, which drove the ChiNext Index up nearly 50% [2] Market Performance Summary - The performance of major indices in 2025 included: - Shanghai Composite Index: 18.41% increase in January, with a total of 8.74% in September [3] - ChiNext Index: 49.57% increase in January, but a notable drop of over 7% in April due to external shocks [3] - Shenzhen Component Index: 29.87% increase in January, with fluctuations throughout the year [3] - Hang Seng Index: 27.77% increase in January, with varying monthly performance [3] Private Fund Performance Summary - In the private fund market, 2025 saw a total of 5,139 products with data showing an average cumulative return of 31.83%, with only 229 products (4.46%) achieving positive returns for 12 consecutive months [4] - Among the top-performing private funds, 1,006 products had performance data, with 88 (8.75%) achieving positive returns for 12 consecutive months [4] - The top 20 products in the private fund sector had a threshold for cumulative returns, with a majority being quantitative long products [4] Notable Products and Managers - The "Square and Peak Zhongzheng 2000 Index Enhanced 21" managed by Square and Investment achieved significant returns, with the highest monthly return in August [7] - The "Yizu Qinggui Li Dong" managed by Yizu Investment topped the list among private funds under 5 billion, with substantial returns in March and December [10] - The "Hua Nian Progress 2A" managed by Xue Yuxin ranked fifth, showcasing strong performance [11] Excess Return Analysis - In 2025, 802 quantitative long products had data on excess returns, with only 30 (3.74%) achieving positive excess returns for 12 consecutive months [13] - The top 10 products in terms of cumulative excess returns were managed by firms such as Jintong Investment and Ming Stone Fund [14][16]

Core Viewpoint - The report highlights the performance of various market factors and investment strategies over the week of January 5 to January 9, 2026, indicating a mixed performance across different factors and sectors, with notable trends in momentum and valuation factors [4][5][6]. Factor Performance - Major factors such as beta, residual volatility, and size factors yielded positive returns of 1.07%, 1.02%, and 0.59% respectively, while the momentum factor showed a significant negative return of -1.08% [4]. - In the CSI 300 stock pool, the best-performing factors included 5-day average turnover rate (4.90%), relative turnover volatility (4.59%), and quarterly revenue growth rate (3.92%), while the worst performers were momentum-adjusted large orders (-1.11%), ROA stability (-1.15%), and ROE stability (-1.43%) [5]. - In the CSI 500 stock pool, the top factors were gross margin TTM (1.29%), quarterly net profit growth rate (1.09%), and total asset growth rate (0.81%), with the worst being price-to-book ratio (-3.51%), TTM price-to-earnings ratio inverse (-4.06%), and price-to-earnings ratio (-4.69%) [5]. - In the liquidity 1500 stock pool, the best factors were gross margin TTM (2.17%), quarterly revenue growth rate (2.14%), and quarterly operating profit growth rate (1.85%), while the worst were the correlation of intraday volatility with transaction amount (-2.64%), price-to-earnings ratio (-3.01%), and TTM price-to-earnings ratio inverse (-3.18%) [5]. Industry Factor Performance - The net asset growth rate factor performed well in the non-bank financial and diversified sectors, while the net profit growth rate factor excelled in the diversified sector [6]. - The per-share net asset factor showed strong performance in the real estate and beauty care sectors, and the per-share operating profit TTM factor performed well in the diversified sector [6]. - The 5-day momentum factor exhibited momentum effects in media, communication, steel, and pharmaceutical sectors, while showing reversal effects in coal and agriculture sectors [6]. - Valuation factors like BP performed well in real estate and leisure services, while EP performed well in banking and non-bank financial sectors [7]. Investment Strategy Performance - The PB-ROE-50 combination achieved significant excess returns in the CSI 800 and overall market stock pools, with excess returns of 1.36% in the CSI 800 and 1.23% in the overall market, but a negative excess return of -2.18% in the CSI 500 stock pool [8]. - The private equity research tracking strategy generated positive excess returns, while the public equity research stock selection strategy had a relative excess return of -0.31% compared to the CSI 800 [9]. - The block trading combination achieved an excess return of 0.69% relative to the CSI All Index [10]. - The targeted issuance combination experienced a pullback in excess returns, with a relative excess return of -1.58% compared to the CSI All Index [11].

Core Insights - The report presents a systematic framework for automated factor research based on Large Language Models (LLM) and Prompt Engineering, aiming to explore the potential applications of AI in the entire quantitative investment chain [1] - The framework was first applied to low-frequency price-volume factors, optimizing the classic Alpha158 factor library and transitioning from an "optimization" paradigm to a "generation" paradigm [1] - AI demonstrated strong factor discovery capabilities in both fundamental and high-frequency data domains, successfully generating new factors and enhancing traditional factor libraries [1] - The report also explores AI's application in unstructured text analysis, utilizing the Gemini model to interpret sentiment from extensive research memos, creating unique sentiment indicators that effectively integrate into stock selection strategies [1] Group 1: Low-Frequency Price-Volume Factor Optimization - The framework was initially applied to the optimization of low-frequency price-volume factors, using the Alpha158 factor library as a foundation for optimization experiments [1] - AI identified logical flaws in original factors and proposed effective improvements, with optimization effects being consistent across multiple time windows from 5 to 60 days [1] - New factors generated by AI, with low correlation to sample factors, showed robust out-of-sample performance, with some factors achieving an Information Coefficient Information Ratio (ICIR) above 1.0 [1] Group 2: Fundamental and High-Frequency Factor Discovery - In the fundamental dimension, AI not only generated enhanced versions of classic factors but also innovatively expanded value, quality, and growth factors from novel perspectives [1] - In the high-frequency dimension, AI was empowered to directly generate Python code, uncovering a set of novel and high-performing high-frequency factors, with some strong signal factors achieving annualized returns exceeding 60% [1] - Integrating the AI-generated high-frequency factor library into the AGRU neural network model significantly improved annualized excess returns from 18.24% to 25.28% [1] Group 3: Alternative Data Processing and Sentiment Analysis - The report investigates AI's potential in processing alternative data, analyzing nearly one million words of research memos using the Gemini 2.5 Pro model [1] - A weekly sentiment factor was constructed, revealing unique asymmetric predictive capabilities, where negative sentiment strongly predicted future price declines, achieving annualized excess returns of 8.26% [1] - This sentiment factor exhibited low correlation with traditional price-volume and fundamental factors, serving as an independent and effective supplementary information source [1] Group 4: Comprehensive Strategy Development - A multi-dimensional information fusion strategy was developed, integrating AI-discovered high-frequency factors with low-frequency market data into the AGRU neural network to form a core Alpha [1] - The final strategy, enhanced by AI sentiment factors for risk adjustment, improved annualized excess returns from 11.15% to 11.81% while maintaining turnover rates [1] - The strategy demonstrated a significant increase in the information ratio from 2.18 to 2.31, validating AI's potential to empower quantitative research across multiple stages and achieve a "1+1>2" effect [1]