Quantitative Models and Construction Methods - **Model Name**: Hong Kong Stock Selection Portfolio Strategy **Model Construction Idea**: The strategy is based on a dual-layer selection process that combines fundamental and technical analysis to identify outperforming stocks from an analyst-recommended stock pool[14][15] **Model Construction Process**: 1. **Analyst Recommendation Pool**: Constructed using three types of analyst recommendation events: upward earnings revisions, first-time coverage, and research reports with unexpected positive titles[15] 2. **Fundamental and Technical Screening**: Stocks in the recommendation pool are further filtered based on fundamental support and technical resonance to identify stocks with both strong fundamentals and positive technical trends[15] 3. **Backtesting**: The backtesting period spans from January 1, 2010, to June 30, 2025, assuming a fully invested portfolio with transaction costs considered[15] **Model Evaluation**: The strategy demonstrates strong performance with significant excess returns over the Hang Seng Index[15] - **Model Name**: Stable New High Stock Screening **Model Construction Idea**: This model leverages momentum and trend-following strategies, focusing on stocks that have recently reached 250-day highs and exhibit stable price paths[20][22] **Model Construction Process**: 1. **250-Day High Distance Calculation**: $ 250\text{-day high distance} = 1 - \frac{\text{Close}_{\text{latest}}}{\text{ts\_max(Close, 250)}} $ Where $\text{Close}_{\text{latest}}$ is the latest closing price, and $\text{ts\_max(Close, 250)}$ is the maximum closing price over the past 250 trading days[22] 2. **Screening Criteria**: - Stocks must have reached a 250-day high in the past 20 trading days - Analyst coverage: At least five "Buy" or "Overweight" ratings in the past six months - Relative strength: Top 20% in 250-day returns among all Hong Kong stocks - Stability: Evaluated using metrics such as price path smoothness and the time-series average of the 250-day high distance over the past 120 days[22][23] 3. **Final Selection**: The top 50 stocks based on stability and trend continuation metrics are selected[23] **Model Evaluation**: The model effectively identifies stocks with strong momentum and stable price trends, aligning with the principles of momentum investing[20][22] Model Backtesting Results - **Hong Kong Stock Selection Portfolio Strategy**: - Annualized Return: 19.11% - Excess Return over Hang Seng Index: 18.48% - Information Ratio (IR): 1.22 - Maximum Drawdown: 23.73%[15][19] - **Stable New High Stock Screening**: - Not explicitly quantified in the report, but the model identifies stocks with strong recent performance and stable price paths, such as those in the financial, healthcare, and consumer sectors[22][23] Quantitative Factors and Construction Methods - **Factor Name**: 250-Day High Distance **Factor Construction Idea**: Measures the proximity of the latest closing price to the highest closing price in the past 250 trading days, capturing momentum and trend-following characteristics[22] **Factor Construction Process**: $ 250\text{-day high distance} = 1 - \frac{\text{Close}_{\text{latest}}}{\text{ts\_max(Close, 250)}} $ - If the latest closing price reaches a new high, the factor value is 0 - If the price has fallen from the high, the factor value is positive, indicating the degree of pullback[22] **Factor Evaluation**: This factor is effective in identifying stocks with strong momentum and limited pullbacks, which are likely to continue their upward trends[22] Factor Backtesting Results - **250-Day High Distance**: - Specific performance metrics are not provided, but the factor is used to screen stocks with strong momentum and stable trends, contributing to the selection of outperforming stocks in the financial, healthcare, and consumer sectors[22][23]

Group 1: A-shares and Hong Kong Stocks - The A-share market is expected to show a volatile adjustment pattern in July, with short-term momentum effects possibly leading to continued increases, followed by a potential adjustment phase [4][30] - The Hong Kong stock market is anticipated to align with the A-share market's overall rhythm, but the A-share's chip structure is superior, and the Hang Seng AH premium index is reversing from a low position, reducing the attractiveness of Hong Kong stocks [4][30] - In early July, the growth style is expected to outperform, while dividend stocks may experience relative volatility; however, as momentum effects fade and tariff policy uncertainties increase in mid to late July, growth style may face headwinds, allowing dividend style to shine [4][30] Group 2: US Stocks and Gold - The risk trend model indicates that the risk level of US stocks has reached a high point, predicting a volatile trend in July, with the expiration of the tariff suspension period on July 9 likely impacting the market [4][30] - The gold market is assessed to have a moderate risk level, with no significant overvaluation or undervaluation; expectations for interest rate cuts are rising, leading to a gradual strengthening of the market [4][30] - Overall, US stocks and gold are expected to maintain a reverse volatile pattern, awaiting catalysts from geopolitical events, policy changes, and US economic data releases [4][30] Group 3: Government Bonds and US Treasuries - The government bond market is supported by a slow economic recovery, maintaining confidence in policy easing, with liquidity improvement expectations becoming clearer post-quarter [4][30] - The US Treasury market is influenced by external uncertainties that elevate risk aversion, supporting a downward trend in interest rates, although supply pressures and policy fluctuations limit the extent of this decline [4][30] - The overall interest rate trend is expected to show a downward movement, influenced by domestic recovery and flexible policies alongside persistent US inflation and debt supply [4][30] Group 4: Fund Allocation Recommendations - A balanced allocation strategy is recommended, anticipating that the market may exhibit a volatile adjustment trend in the future, suggesting a wait-and-see approach for optimal timing [4][30]

Investment Rating - The report maintains an "Overweight" investment rating for the financial products industry [1]. Core Insights - The report emphasizes the importance of momentum-driven technical analysis and risk management in the financial products sector, highlighting the dual framework of momentum and risk control [1][2]. Summary by Sections 1. Momentum-Driven Technical Analysis - The report discusses the momentum effect, indicating that assets generally exhibit a certain degree of trend persistence in the short term [13]. - Moving averages (MA) are defined as a key indicator for assessing momentum, helping to smooth price fluctuations and identify potential trends [14][15]. - The report categorizes moving averages into short-term, medium-term, and long-term, each serving different market analysis needs [16]. - The MACD (Moving Average Convergence Divergence) is introduced as a significant technical analysis tool, consisting of the DIF line, DEA line, and MACD histogram, which helps investors grasp market trends [25][26]. - The report also covers the JAX (Jian Line) indicator, which combines price and volume data to assess medium to long-term trends and trading opportunities [34][35]. 2. Risk Management - The report introduces the Risk Degree Indicator (TR), which evaluates the relative position of assets in terms of risk, considering both spatial and temporal dimensions [45]. - The TR indicator operates within a normal range of 0 to 100, with higher values indicating increased investment risk and lower values suggesting reduced risk [50]. - Statistical analysis shows that the TR indicator effectively identifies local tops and bottoms in the A-share market, with a 45.74% probability of local lows occurring when TR is below 20 [50][51]. 3. Risk Trend Model - The report outlines a risk trend model that incorporates both momentum and risk dimensions to score assets, aiding in the development of timing strategies [44][49]. - It emphasizes the need for dynamic adjustments in parameters based on market conditions to enhance the accuracy of the JAX indicator [41][42]. 4. Timing Strategies - The report suggests constructing timing strategies based on historical data, with a focus on both periodic and non-periodic models to improve predictive effectiveness [44][49].

这一年，D.E. Shaw 为计算机行业做了两个贡献。一个副总裁带队，做出了当时罕见的免费电子邮件产 品 Juno，成功上市；另一个副总裁离职，带着自己和老板讨论产生的好点子开车去了西雅图，做出了全 世界的电商鼻祖、市值超过 20000 亿美元的亚马逊。 30 年后，又有一家量化公司的 "副业" 影响整个计算机行业：管理数百亿元的中国头部量化公司幻方， 推出大语言模型 DeepSeek R1，没花一分钱营销就震撼全球，用户涌来的速度甚至快过早年的抖音。 贝索斯创办亚马逊，或者梁文锋造出 DeepSeek 的主要原因自然不是因为他们做过量化，而是因为他们 骨子里都是创业者。但量化投资这个极度追求人才密度且极度保密的行业文化，确实提供了适合大模型 研发的环境。 招来一群聪明人不必然导致创新，叠加一个简单的环境才够。量化公司证明了这一点，DeepSeek 则证明 这也适用于大模型研发。 剥离主观因素，在数据里挖掘规律 从十万次交易到千亿参数的 AI 进化。 文 丨 孙海宁 编辑 丨 黄俊杰 1994 年，量化公司是当时最神秘最热门的技术公司，他们雇用数学家和物理学家，成批买来高性能计算 机做交易。这个行业里的标杆公 ...

Quantitative Models and Factor Construction Multi-Horizon Factor - **Factor Name**: Multi-Horizon Factor - **Construction Idea**: This factor captures short-term reversal, medium-term momentum, and long-term reversal effects by analyzing moving average (MA) data across multiple time horizons [2][14][21] - **Construction Process**: - Calculate moving averages for different time horizons \( L = [3, 5, 10, 20, 30, 60, 90, 120, 180, 240, 270, 300] \) using the formula: \[ A_{j t,L} = \frac{P_{j,\,d-L+1}^{t} + \cdots + P_{j,d}^{t}}{L} \] where \( P_{j,d}^t \) represents the price of stock \( j \) at time \( t \) [21] - Standardize the moving average factor: \[ \tilde{A}_{j t,\,L} = \frac{A_{j t,\,L}}{P_{j}^{t}} \] [22] - Perform cross-sectional regression of stock returns on lagged standardized moving average factors: \[ r_{j,t} = \beta_{0,t} + \Sigma_{i}\beta_{i,t}\tilde{A}_{j t-1,L_{i}} + \epsilon_{j,t} \] [23] - Predict next-period regression coefficients by averaging the past 25 weeks' coefficients: \[ E\left[\beta_{i,\,t+1}\right] = \frac{1}{25}\,\sum_{m=1}^{25}\,\beta_{i,t+1-m} \] [24] - Use predicted coefficients and new factor values to estimate next-period returns: \[ E\left[r_{j,t+1}\right] = \Sigma_{i}\,E\left[\beta_{i,\,t+1}\right]\tilde{A}_{j t,\,L_{i}} \] [25] - Rank stocks by predicted returns and construct long-short portfolios [26] - **Evaluation**: The factor demonstrates strong predictive power for stock returns across different market segments, with positive IC values dominating [30][32] LLT Trend Factor - **Factor Name**: LLT Trend Factor - **Construction Idea**: To address the lagging sensitivity of MA, the LLT (Low-Lag Trendline) indicator is used as a replacement. LLT reduces delay and better captures momentum and reversal effects [14][76] - **Construction Process**: - LLT is calculated using a second-order linear filter with the recursive formula: \[ LLT = \begin{cases} P(T), & T=1,2 \\ (2-2\alpha)LLT(T-1) - (1-\alpha)^2LLT(T-2) + \left(\alpha-\frac{\alpha^2}{4}\right)P(T) \\ + \left(\frac{\alpha^2}{2}\right)P(T-1) - \left(\alpha-\frac{3}{4}\alpha^2\right)P(T-2), & \text{else} \end{cases} \] where \( \alpha = \frac{2}{1+N} \) and \( N \) is the smoothing parameter [76] - Replace MA with LLT in the multi-horizon factor construction process [76] - **Evaluation**: LLT-based factors outperform MA-based factors in terms of IC mean, positive IC ratio, and predictive power for asset returns [82][84] --- Backtesting Results Multi-Horizon Factor - **Annualized Return**: 25.40% [3][48] - **Annualized Volatility**: 14.12% [48] - **Maximum Drawdown**: 13.31% [48] - **IR**: 1.81 [48] LLT Trend Factor - **Annualized Return**: 29.58% [4][103] - **Annualized Volatility**: 10.46% [103] - **Maximum Drawdown**: 11.57% [103] - **IR**: 2.51 [103]