Quantitative Models and Construction Methods 1. Model Name: Dividend Impact Prediction Model - **Model Construction Idea**: The model aims to predict the impact of dividends on index futures pricing by estimating the dividend points for each contract and incorporating them into the theoretical pricing framework[7][10][20] - **Model Construction Process**: 1. **Estimate Component Stocks' Net Profit**: Use annual reports, quick reports, earnings warnings, or analysts' forecasts to estimate net profits[23][24] 2. **Calculate Pre-Tax Total Dividends**: Based on the assumption that the dividend payout ratio remains unchanged, calculate the total dividend amount as: $$ \text{Estimated Dividend Amount} = \text{Estimated Net Profit} \times \text{Dividend Payout Ratio} $$ If no dividends were distributed in the previous year, assume no dividends this year[28] 3. **Calculate Dividend Impact on Index**: - Dividend Yield: $$ \text{Dividend Yield} = \frac{\text{Tax-Adjusted Total Dividend}}{\text{Latest Market Value}} $$ - Dividend Points: $$ \text{Dividend Points Impact} = \text{Stock Weight} \times \text{Dividend Yield} $$ - Adjust stock weights using the formula: $$ w_{it} = \frac{w_{i0} \times (1 + R)}{\sum_{1}^{n} w_{i0} \times (1 + R)} $$ where \( w_{i0} \) is the initial weight, and \( R \) is the return over the period[25] 4. **Predict Contract Impact**: Aggregate all dividend impacts before the contract's delivery date to estimate the total dividend effect on the futures contract[30] 5. **Theoretical Pricing**: - For discrete dividends: $$ F_t = (S_t - D)(1 + r) $$ where \( D \) is the present value of dividends, and \( r \) is the risk-free rate[33] - For continuous dividends: $$ F_t = S_t e^{(r-d)(T-t)} $$ where \( d \) is the annualized dividend yield[34] - **Model Evaluation**: The model provides a systematic approach to incorporate dividend forecasts into futures pricing, enhancing accuracy in predicting contract price movements[7][10][20] --- Model Backtesting Results 1. Dividend Impact Prediction Model - **Dividend Points for June Contracts**: - SSE 50: 17.28 - CSI 300: 20.75 - CSI 500: 35.79 - CSI 1000: 32.06[7][10][15] - **Annualized Hedging Costs (Excluding Dividends)**: - SSE 50: 0.76% - CSI 300: 5.14% - CSI 500: 12.79% - CSI 1000: 18.63%[7][10][15] - **Remaining Impact of Dividends on June Contracts**: - SSE 50: 0.64% - CSI 300: 0.53% - CSI 500: 0.63% - CSI 1000: 0.54%[15]

Core Insights - The report emphasizes the importance of benchmark anchoring for generating excess returns in stock portfolios, suggesting that fund managers should focus on individual stock alpha while controlling style and sector deviations [4][5][6]. Financial Engineering - **Strategy 1: Core-Satellite Approach**: Allocate W% of the portfolio to benchmark anchoring and (1-W%) to active management, allowing for better tracking error control while maintaining excess returns. A suggested W parameter is 40% for specific performance metrics [4]. - **Strategy 2: Industry Neutrality**: Ensure the stock portfolio's industry allocation matches that of the benchmark (CSI 300), which can reduce tracking error and lower the probability of underperformance by over 10% compared to the benchmark [5]. - **Strategy 3: Style Neutrality**: Maintain the original stock selection but adjust weights to minimize style deviation from the benchmark, which can effectively lower tracking error at minimal cost [6]. - **Strategy 4: Barbell Strategy**: For funds with distinct style biases, a dual strategy combining growth and defensive investments can help reduce tracking error and volatility, suitable for long-term investment goals [6]. Steel Industry - The report discusses the cyclical nature of national debt cycles, categorizing them into three phases: local government debt, centralization of local debt, and monetization of national debt, reflecting the broader economic cycles of labor and wealth [7]. Electronics Industry - **Company Overview**: 纳芯微 (Naxin Micro) is a leading player in automotive analog chips, with a product portfolio that includes over 3,300 models. The company holds the top market share among domestic manufacturers in automotive analog chips and magnetic sensors [8]. - **Financial Performance**: The company expects significant revenue growth, projecting revenues of 29.59 billion, 37.95 billion, and 47.29 billion yuan for 2025-2027, with corresponding net profits of -0.81 billion, 1.03 billion, and 2.95 billion yuan [8]. Pharmaceutical Industry - **Company Strategy**: 阳光诺和 (Sunshine Novo) plans to acquire 100% of 朗研 (Langyan) to accelerate innovation and enhance its business ecosystem, focusing on R&D services, pipeline cultivation, and a new quality industrial chain [10]. - **Financial Projections**: The company anticipates net profits of 2.33 billion, 2.88 billion, and 3.55 billion yuan for 2025-2027, reflecting growth rates of 31.3%, 23.8%, and 23.0% respectively [10]. Retail Industry - **Market Overview**: The retail sector showed a year-on-year growth of 5.1% in April, indicating a stable recovery with some sub-sectors improving. Key players include 华住集团 (Huazhu Group) and 永辉超市 (Yonghui Supermarket) [15]. - **Investment Opportunities**: The report highlights potential in sectors benefiting from tourism and new retail formats, suggesting a positive outlook for companies adapting to changing consumer behaviors [15]. Textile and Apparel Industry - **Company Performance**: 滔搏 (Tao Bo) reported a revenue decline of 6.6% for FY2025, with a significant drop in net profit by 41.9%, attributed to a challenging consumer environment and inventory adjustments [16]. - **Future Outlook**: Despite short-term pressures, the company is expected to recover with projected net profits of 13.01 billion, 14.81 billion, and 16.47 billion yuan for FY2026-2028 [16]. Food and Beverage Industry - **Company Strategy**: 青岛啤酒 (Qingdao Beer) is focusing on market expansion during peak seasons, leveraging cost advantages and scale effects to enhance profitability [18]. - **Financial Forecast**: The company projects net profits of 48.1 billion, 52.1 billion, and 56.5 billion yuan for 2025-2027, with growth rates of 10.7%, 8.2%, and 8.6% respectively [18]. Snack Industry - **Company Development**: 三只松鼠 (Three Squirrels) is expanding its product categories and distribution channels, aiming to create a comprehensive supply chain that integrates manufacturing, branding, and retail [21]. - **Market Positioning**: The company is leveraging its efficient supply chain to tap into broader market opportunities, transitioning from online to offline sales and exploring new retail formats [21].

Core Insights - The report presents a novel investment portfolio construction method that combines stock price momentum with hierarchical clustering (HC) to address the instability and concentration issues of Markowitz mean-variance (MV) portfolio weights [2][22] - The proposed Hierarchical Momentum (HM) strategy shows potential applicability in various domains such as stock portfolio construction, ETF portfolio construction, and asset allocation in the domestic market [2][22] Hierarchical Momentum Strategy - The HM strategy derives a distance function from the Pearson correlation coefficients between asset returns, using a bottom-up recursive approach to cluster assets based on proximity, resulting in a dendrogram [3][24] - At a certain height in the dendrogram, a horizontal cut is made to divide the tree into n clusters, identifying the assets with the highest momentum scores within each cluster while assigning zero weight to those with negative momentum scores [3][24] Empirical Results - The backtesting period spans from June 1997 to August 2022, utilizing a high-dimensional dataset from the MSCI All Country World Index (ACWI), which includes large-cap and mid-cap stocks from 23 developed and 24 emerging markets [5][43] - After accounting for transaction costs, the HM strategy outperforms all other strategies in terms of cumulative returns, average returns, risk-adjusted returns (Sharpe and Sortino ratios), and risk metrics (volatility and maximum drawdown) [5][55] - The HM strategy demonstrates improved stability in industry allocation compared to the Maximum Momentum (MM) and Threshold Momentum (TM) strategies, which are known for their potential large drawdown issues [5][56] Methodology - The HM portfolio construction method does not require the inversion of the covariance matrix, instead relying on a hierarchical clustering approach to reduce dimensionality and ensure sparse diversification [24][68] - The method involves two main steps: applying hierarchical clustering to create a distance matrix and then constructing portfolio weights based on the hierarchical structure and momentum scores [24][38] Conclusion - The report emphasizes the importance of sparse diversification in constructing superior investment portfolios, particularly in high-dimensional environments where traditional methods may underperform [68][69] - The HM strategy effectively captures momentum premiums while mitigating risks associated with traditional momentum strategies, demonstrating its robustness across different economic conditions [68][69]

Group 1 - The report introduces a quantitative model solution for addressing the issue of value and growth style switching, based on the combination of odds and win rates [1][8] - The recent performance of the growth style portfolio was 0.82%, while the value style portfolio achieved a return of 1.15% [1][8] Group 2 - The estimated odds for the growth style is 1.09, and for the value style, it is 1.08, indicating a negative correlation between relative valuation levels and expected odds [2][14] - The current win rates indicate that 4 out of 7 indicators favor growth, resulting in a win rate of 58.26% for growth and 41.74% for value [3][18] Group 3 - The latest investment expectation for the growth style is calculated to be 0.22, while the value style has an investment expectation of -0.13, leading to a recommendation for the growth style [4][19] - Since 2013, the annualized return of the style rotation model based on investment expectations is 27.18%, with a Sharpe ratio of 0.98 [4][20]

Quantitative Models and Construction Methods - **Model Name**: Theoretical Pricing Model for Stock Index Futures **Model Construction Idea**: This model aims to calculate the theoretical price of stock index futures by considering the impact of dividends and risk-free interest rates under no-arbitrage conditions [35][36] **Model Construction Process**: 1. **Discrete Dividend Distribution**: - Assume the futures price at time \( t \) is \( F_t \), the spot price is \( S_t \), and the futures contract expires at \( T \). The present value of dividends during \( T-t \) is \( D \), and the risk-free rate during \( T-t \) is \( r \). - If there are \( m \) dividend payments at times \( t_1, t_2, ..., t_m \), with amounts \( D_1, D_2, ..., D_m \), the present value of dividends is: $$ \mathbf{D} = \sum_{\mathrm{i=1}}^{\mathrm{m}} \mathbf{D}_{\mathrm{i}} / (1 + \phi) $$ where \( \phi \) is the risk-free rate between two dividend payments. - The theoretical futures price is: $$ F_t = (S_t - D)(1 + r) $$ [35] 2. **Continuous Dividend Distribution**: - When dividends are distributed continuously, the model assumes the annualized dividend yield is \( d \), and the annualized risk-free rate is \( r \). The theoretical futures price is: $$ F_t = S_t e^{(r-d)(T-t)} $$ [36] Quantitative Factors and Construction Methods - **Factor Name**: Dividend Impact Factor **Factor Construction Idea**: This factor estimates the impact of dividends on stock index futures pricing by predicting the dividend points for index components and their contribution to the index [12][27] **Factor Construction Process**: 1. **Estimate Net Profit**: Use available financial data in the following order of priority: annual reports, quick reports, earnings warnings, trailing twelve-month (TTM) net profit, or analysts' forecasts [27][31] 2. **Calculate Total Dividends**: Assume the dividend payout ratio remains constant for companies with historical dividends. For companies with no prior dividends or negative profits, assume zero dividends [31] 3. **Calculate Dividend Impact on Index**: - Dividend yield: \( \text{Tax-adjusted dividends} / \text{Latest market cap} \) - Dividend points: \( \text{Stock weight} \times \text{Dividend yield} \) - Adjust stock weights using the formula: $$ w_{it} = \frac{w_{i0} \times (1 + R)}{\sum_{1}^{n} w_{i0} \times (1 + R)} $$ where \( w_{i0} \) is the initial weight, and \( R \) is the stock's return [29] 4. **Predict Impact on Futures Contracts**: Aggregate the dividend points for all components before the contract's settlement date [33] Model Backtesting Results - **Theoretical Pricing Model**: - Annualized hedging costs (excluding dividends) for May contracts: - SSE 50: 0.27% - CSI 300: 7.07% - CSI 500: 15.59% - CSI 1000: 18.88% [12][13][15][16] Factor Backtesting Results - **Dividend Impact Factor**: - Remaining impact of dividends on May contracts: - SSE 50: 0.01% - CSI 300: 0.02% - CSI 500: 0.04% - CSI 1000: 0.06% [17]

证券研究报告·金融工程·金工定期报告 金工定期报告 20250507 估值异常因子绩效月报 20250430 2025 年 05 月 07 日 [Table_Tag] [Table_Summary] 报告要点 ◼ 估值偏离 EPD 因子多空对冲绩效（全市场）：2010 年 2 月至 2025 年 4 月，估值偏离 EPD 因子在全体 A 股（剔除北交所股票）中，5 分组多 空对冲的年化收益为 17.65%，年化波动为 10.02%，信息比率为 1.76， 月度胜率为 71.04%，月度最大回撤为 8.93%。 ◼ 缓慢偏离 EPDS 因子多空对冲绩效（全市场）：2010 年 2 月至 2025 年 4 月，缓慢偏离 EPDS 因子在全体 A 股（剔除北交所股票）中，5 分组 多空对冲的年化收益为 16.31%，年化波动为 5.73%，信息比率为 2.85， 月度胜率为 79.23%，月度最大回撤为 3.10%。 ◼ 估值异常 EPA 因子多空对冲绩效（全市场）：2010 年 2 月至 2025 年 4 月，估值异常 EPA 因子在全体 A 股（剔除北交所股票）中，5 分组多空 对冲的年化收益为 17.30%， ...

盈利预期期限结构因子历史表现良好，走势稳定性高。 将盈利预期期限结构"动量 ff"因子与传统分析师预期 调升因子合成后，得到的"合成动量 ff"因子走势进一步改 善，兼具高收益与稳定性。 ► 选股组合表现 在沪深 300、中证 500、中证 800、中证 1000 内分别选择 "合成动量 ff"因子值排名前 50、50、100、100 名的股票， 构成选股组合，组合历史表现良好。 证券研究报告|金融工程研究报告 [Table_Date] 2025 年 5 月 2 日 [Table_Title] 前四个月全部组合跑赢基准——盈利预期期限结构选股月报 202505 [Table_Summary] ► 盈利预期期限结构因子 分析师在某一时点会对上市公司未来多年的盈利做出预 测，我们将预期盈利随未来年度变化的趋势称为盈利预期期 限结构。 2025 年 4 月，沪深 300 选股组合、中证 800 选股组合跑 输基准，超额收益分别为-0.57%、-0.02%；中证 500 选股组 合、中证 1000 选股组合跑赢基准，超额收益分别为 0.73%、 1.13%。 2025 年前 4 个月，沪深 300、中证 500、中证 ...

Group 1: Davis Double-Click Strategy - The Davis Double-Click strategy involves buying stocks with growth potential at a lower price-to-earnings (PE) ratio, waiting for growth to manifest, and then selling for a multiplier effect, achieving a dual benefit from earnings per share (EPS) and PE [1][7] - The strategy has achieved an annualized return of 26.45% during the backtest period from 2010 to 2017, exceeding the benchmark by 21.08% [9] - Year-to-date, the strategy has a cumulative absolute return of 6.18%, outperforming the CSI 500 index by 7.90%, with a weekly excess return of 2.27% [10] Group 2: Net Profit Discontinuity Strategy - The Net Profit Discontinuity strategy focuses on selecting stocks based on fundamental and technical resonance, where "net profit" refers to earnings surprises, and "discontinuity" indicates a significant upward price gap on the first trading day after earnings announcements [2][12] - This strategy has achieved an annualized return of 28.17% since 2010, with an annualized excess return of 26.67% over the benchmark [15] - The current year's cumulative absolute return for this strategy is 8.31%, exceeding the benchmark by 10.03%, with a weekly excess return of 4.31% [15] Group 3: Enhanced CSI 300 Portfolio - The Enhanced CSI 300 portfolio is constructed based on investor preferences, including GARP (Growth at a Reasonable Price), growth, and value investing styles, utilizing PBROE and PEG factors to identify undervalued stocks with strong profitability and growth potential [3][17] - The portfolio has shown stable excess returns in historical backtesting, with a year-to-date excess return of 7.41% relative to the CSI 300 index [17] - The weekly excess return for this portfolio is 1.33%, and the monthly excess return is 4.08% [17]

Quantitative Models and Factor Analysis Dividend Impact Prediction Model - **Model Name**: Dividend Impact Prediction Model - **Model Construction Idea**: The model aims to predict the impact of dividends on stock index futures contracts by estimating the dividend points and their influence on futures pricing. It incorporates historical dividend patterns, company financial reports, and market assumptions to calculate the theoretical impact of dividends on futures contracts[9][22][25] - **Model Construction Process**: 1. **Estimate Component Stocks' Net Profit**: Use annual reports, financial forecasts, and analysts' predictions to estimate the net profit of index component stocks[25][26] 2. **Calculate Pre-Tax Dividend Total**: Based on the assumption that the dividend payout ratio remains unchanged, calculate the total dividend amount using the formula: $$ \text{Estimated Dividend Total} = \text{Net Profit} \times \text{Dividend Payout Ratio} $$ If no dividends were issued in the previous year, assume no dividends for the current year[29] 3. **Calculate Dividend Impact on Index**: - **Dividend Yield**: $$ \text{Dividend Yield} = \frac{\text{Tax-Adjusted Dividend Total}}{\text{Latest Market Value}} $$ - **Dividend Points**: $$ \text{Dividend Points Impact} = \text{Stock Weight} \times \text{Dividend Yield} $$ Stock weights are adjusted using the formula: $$ \mathrm{w_{it}={\frac{w_{i0}\times\ (\ 1+R\ )}{\sum_{1}^{n}w_{i0}\times\ (\ 1+R\ )}}} $$ where \( R \) is the stock's price change over time[27] 4. **Predict Impact on Futures Contracts**: Aggregate the dividend points for all component stocks before the contract's settlement date to estimate the total dividend impact on the futures contract[31] - **Model Evaluation**: The model is robust in incorporating historical data and market assumptions, but its accuracy depends on the reliability of financial forecasts and the stability of dividend payout ratios[9][25][31] --- Backtesting Results of the Model Dividend Impact on Futures Contracts (May 2025 Contracts) - **Shanghai 50 Index Futures (IH)**: - Dividend Points: 0.43 - Remaining Impact: 0.02% - Annualized Hedging Cost (365 days): 2.22% - Annualized Hedging Cost (243 days): 2.58%[10][14] - **CSI 300 Index Futures (IF)**: - Dividend Points: 5.76 - Remaining Impact: 0.15% - Annualized Hedging Cost (365 days): 2.95% - Annualized Hedging Cost (243 days): 3.44%[10][11] - **CSI 500 Index Futures (IC)**: - Dividend Points: 3.49 - Remaining Impact: 0.06% - Annualized Hedging Cost (365 days): 12.82% - Annualized Hedging Cost (243 days): 14.93%[10][12] - **CSI 1000 Index Futures (IM)**: - Dividend Points: 5.93 - Remaining Impact: 0.10% - Annualized Hedging Cost (365 days): 14.22% - Annualized Hedging Cost (243 days): 16.56%[10][13] --- Theoretical Pricing Model for Futures Discrete Dividend Distribution Model - **Model Construction Process**: - Assumes discrete dividend payments at specific time points during the contract period - Formula: $$ \mathbf{D}=\sum_{\mathrm{i=1}}^{\mathrm{m}}\mathbf{D}_{\mathrm{i}}\,/(1+\phi) $$ where \( \phi \) is the risk-free rate between two dividend payment dates - Futures Pricing Formula: $$ F_t = (S_t - D)(1 + r) $$ where \( S_t \) is the spot price, \( D \) is the present value of dividends, and \( r \) is the risk-free rate[34] Continuous Dividend Distribution Model - **Model Construction Process**: - Assumes dividends are distributed continuously over the contract period - Formula: $$ F_t = S_t e^{(r-d)(T-t)} $$ where \( d \) is the annualized dividend yield, \( r \) is the risk-free rate, and \( T-t \) is the time to maturity[35]

Quantitative Models and Construction Methods - **Model Name**: Theoretical Pricing Model for Stock Index Futures **Model Construction Idea**: This model calculates the theoretical price of stock index futures by considering the impact of dividends, risk-free interest rates, and the time to maturity. It assumes no arbitrage conditions in the market[34][35] **Model Construction Process**: 1. **Discrete Dividend Distribution**: - Assume the futures price at time \( t \) is \( F_t \), the spot price is \( S_t \), the maturity date is \( T \), and the present value of dividends during \( T-t \) is \( D \). The risk-free rate over \( T-t \) is \( r \). - If there are \( m \) dividend payments at times \( t_1, t_2, ..., t_m \), with amounts \( D_1, D_2, ..., D_m \), the present value of dividends is: $$ \mathbf{D} = \sum_{\mathrm{i=1}}^{\mathrm{m}} \mathbf{D}_{\mathrm{i}} / (1 + \phi) $$ Here, \( \phi \) represents the risk-free rate between two dividend payments. - The theoretical futures price is: $$ F_t = (S_t - D)(1 + r) $$ 2. **Continuous Dividend Distribution**: - When dividends are distributed continuously, the theoretical futures price is: $$ F_t = S_t e^{(r-d)(T-t)} $$ Here, \( d \) is the annualized dividend yield, and \( r \) is the annualized risk-free rate[34][35] Quantitative Factors and Construction Methods - **Factor Name**: Dividend Impact Factor **Factor Construction Idea**: This factor estimates the impact of dividends on stock index futures by predicting the dividend payout of index constituents and their contribution to the index[22][25] **Factor Construction Process**: 1. **Estimate Net Profit**: Use annual reports, earnings forecasts, and other financial data to estimate the net profit of index constituents[25][26] 2. **Calculate Total Dividend**: Assume the dividend payout ratio remains constant. For companies without prior dividends, assume no dividends. For companies with negative profits, set the dividend ratio to zero[29] 3. **Calculate Dividend Impact on Index**: - Dividend Yield = Total Dividend / Latest Market Value - Dividend Impact on Index = Stock Weight × Dividend Yield - Adjust stock weights based on price changes using the formula: $$ \mathrm{w_{it} = \frac{w_{i0} \times (1+R)}{\sum_{1}^{n} w_{i0} \times (1+R)}} $$ Here, \( w_{i0} \) is the initial weight, and \( R \) is the price change[27] 4. **Predict Impact on Futures Contracts**: Aggregate the dividend impact for all constituents before the contract's expiration date[31] Backtesting Results of Models - **Theoretical Pricing Model**: - For the May contracts of major indices, the annualized hedging costs (excluding dividends) are: - CSI 300: 9.63% - CSI 500: 17.67% - CSI 1000: 19.08%[10][11][12] Backtesting Results of Factors - **Dividend Impact Factor**: - Remaining dividend impact on May contracts: - SSE 50: 0.00% - CSI 300: 0.20% - CSI 500: 0.07% - CSI 1000: 0.14%[13]