《微盘股继续领涨市场，扩散指数已 达较高区间——微盘股指数周报 20251114》 - 2025.11.18 《连板情绪持续发酵，GRU 行业轮动调 入 基 础 化 工 — — 行 业 轮 动 周 报 20251109》 - 2025.11.11 《微盘股高位盘整，增长逻辑未改变— — 微 盘 股 指 数 周 报 20251031 》 - 2025.11.03 《上证周中突破 4000 点，扩散指数行 业轮动调入电力设备及新能源——行 业轮动周报 20251102》 - 2025.11.02 《微盘股触发看多信号，看好微盘 10 月 后 续 表 现 — — 微 盘 股 指 数 周 报 20251017》 - 2025.10.22 证券研究报告：金融工程报告 研究所 分析师:黄子崟 SAC 登记编号:S1340523090002 Email:huangziyin@cnpsec.com 研究助理:金晓杰 SAC 登记编号:S1340124100010 Email:jinxiaojie@cnpsec.com 近期研究报告 《上证强于双创调整空间不大，ETF 资 金持续配置金融地产与 TMT 方向——行 业轮动周报 ...

证券研究报告：金融工程报告 发布时间：2025-11-25 研究所 分析师:黄子崟 SAC 登记编号:S1340523090002 Email:huangziyin@cnpsec.com 研究助理:李子凯 SAC 登记编号:S1340124100014 Email:lizikai@cnpsec.com 近期研究报告 《微盘股继续领涨市场，扩散指数已达 较高区间 — — 微盘股指数周报 20251114》 - 2025.11.18 《连板高度打开情绪持续发酵，GRU 行 业轮动调入房地产——行业轮动周报 20251116》 - 2025.11.17 《连板情绪持续发酵，GRU 行业轮动调 入基础化工 — — 行业轮动周报 20251109》– 2025.11.10 《上证周中突破 4000 点，扩散指数行业 轮动调入电力设备及新能源——行业 轮动周报 20251102》 – 2025.11.03 《贵金属回调风偏修复，GRU 行业轮动 调入非银行金融——行业轮动周报 20251028》 – 2025.10.27 《上证强于双创调整空间不大，ETF 资 金持续配置金融地产与 TMT 方向——行 业轮动周 2025 ...

Quantitative Models and Construction GRU Model - **Model Name**: GRU (Generalized Rotation Unit) Model - **Model Construction Idea**: The GRU model is designed to capture industry rotation trends and optimize stock selection by leveraging short-term and long-term market dynamics[3][4][5] - **Model Construction Process**: - The GRU model is applied across different stock pools (e.g., CSI 300, CSI 500, CSI 1000) to evaluate multi-factor performance - It incorporates multiple sub-models such as `barra1d`, `barra5d`, `open1d`, and `close1d` to assess short-term and long-term factor returns - The model evaluates the long-short performance of stocks by ranking them based on factor scores and constructing portfolios with the top 10% (long) and bottom 10% (short) stocks[3][4][5] - **Model Evaluation**: The GRU model demonstrates strong performance in capturing positive long-short returns, particularly in the `barra5d` and `close1d` sub-models, which show consistent strength across different stock pools[4][5][6] --- Quantitative Factors and Construction Style Factors (Barra Factors) - **Factor Names**: Beta, Size, Momentum, Volatility, Non-linear Size, Valuation, Liquidity, Profitability, Growth, Leverage[14][15] - **Factor Construction Ideas**: - These factors are designed to capture specific market characteristics such as risk, size, valuation, and growth potential - They are derived from historical price data, financial metrics, and analyst forecasts - **Factor Construction Process**: - **Beta**: Historical beta - **Size**: Natural logarithm of total market capitalization - **Momentum**: Mean of historical excess return series - **Volatility**: Weighted combination of historical excess return volatility, cumulative excess return deviation, and residual return volatility - **Non-linear Size**: Cubic transformation of size - **Valuation**: Inverse of price-to-book ratio - **Liquidity**: Weighted combination of monthly, quarterly, and annual turnover rates - **Profitability**: Weighted combination of analyst forecasted earnings-to-price ratio, inverse of price-to-cash flow ratio, inverse of trailing twelve-month price-to-earnings ratio, and forecasted growth rates - **Growth**: Weighted combination of earnings growth rate and revenue growth rate - **Leverage**: Weighted combination of market leverage, book leverage, and debt-to-asset ratio[15] - **Factor Evaluation**: Valuation, leverage, and volatility factors showed strong long performance, while beta, momentum, and liquidity factors performed well on the short side during the week[16] Fundamental Factors - **Factor Names**: Static Financial Indicators, Growth Indicators, Surprise Growth Indicators - **Factor Construction Ideas**: These factors are derived from financial statements and are designed to capture company fundamentals such as profitability, growth, and operational efficiency - **Factor Construction Process**: - Financial indicators are calculated using trailing twelve-month (TTM) data - Growth indicators include metrics like revenue growth and earnings growth - Surprise growth indicators measure deviations from analyst expectations[19][20][22] - **Factor Evaluation**: Static financial indicators showed significant negative returns, while growth and surprise growth indicators had mixed performance. Low-growth stocks outperformed across all stock pools[20][22][27] Technical Factors - **Factor Names**: Short-term Momentum, Short-term Volatility, Medium-term Momentum, Medium-term Volatility, Median Absolute Deviation - **Factor Construction Ideas**: These factors are derived from historical price data and are designed to capture price trends and volatility - **Factor Construction Process**: - Momentum factors are calculated as the average excess return over specific time windows (e.g., 20-day, 60-day, 120-day) - Volatility factors are calculated as the standard deviation of returns over specific time windows - Median absolute deviation measures the dispersion of returns around the median[20][24][26] - **Factor Evaluation**: Short-term momentum and volatility factors showed positive returns, while medium-term factors generally underperformed. Low-volatility and low-momentum stocks were favored[20][24][26] --- Model Backtesting Results GRU Model - **Open1d**: Weekly excess return of 1.10%, monthly return of 1.55%, and YTD return of 7.19%[34] - **Close1d**: Weekly excess return of 1.84%, monthly return of 3.56%, and YTD return of 6.24%[34] - **Barra1d**: Weekly excess return of 0.45%, monthly return of -0.26%, and YTD return of 5.19%[34] - **Barra5d**: Weekly excess return of 1.77%, monthly return of 4.51%, and YTD return of 9.23%[34] - **Multi-factor Portfolio**: Weekly excess return of 0.75%, monthly return of 1.16%, and YTD return of 1.65%[34] Style Factors - **Beta**: Weekly return of -5.67%, monthly return of -10.16%, and YTD return of 21.16%[17] - **Momentum**: Weekly return of 4.04%, monthly return of -9.28%, and YTD return of 18.32%[17] - **Liquidity**: Weekly return of -2.91%, monthly return of 6.35%, and YTD return of 11.43%[17] - **Size**: Weekly return of 2.67%, monthly return of 18.45%, and YTD return of 41.09%[17] - **Non-linear Size**: Weekly return of 1.80%, monthly return of -7.67%, and YTD return of 35.55%[17] - **Growth**: Weekly return of 1.59%, monthly return of -2.69%, and YTD return of 2.47%[17] - **Profitability**: Weekly return of 1.13%, monthly return of 0.03%, and YTD return of 15.36%[17] - **Volatility**: Weekly return of 1.35%, monthly return of -1.31%, and YTD return of 4.82%[17] - **Leverage**: Weekly return of 1.36%, monthly return of 3.48%, and YTD return of 15.46%[17] - **Valuation**: Weekly return of 1.45%, monthly return of 4.88%, and YTD return of 2.98%[17] Fundamental Factors - **Net Profit Surprise Growth**: Weekly return of 3.62%, monthly return of -2.63%, and YTD return of 37.43%[23] - **Operating Profit Margin Surprise Growth**: Weekly return of 1.77%, monthly return of 1.65%, and YTD return of 2.46%[23] - **ROA Surprise Growth**: Weekly return of 0.73%, monthly return of 0.19%, and YTD return of 11.22%[23] - **ROA Growth**: Weekly return of -0.70%, monthly return of 0.10%, and YTD return of 25.43%[23] Technical Factors - **20-day Momentum**: Weekly return of 1.88%, monthly return of 5.00%, and YTD return of -5.82%[21][24] - **60-day Momentum**: Weekly return of -10.66%, monthly return of -0.05%, and YTD return of -5.73%[21][24] - **120-day Momentum**: Weekly return of 0.13%, monthly return of 0.13%, and YTD return of -3.53%[21][24] - **20-day Volatility**: Weekly return of 0.08%, monthly return of -0.79%, and YTD return of 11.01%[21][24] - **60-day Volatility**: Weekly return of -0.67%, monthly return of -3.35%, and YTD return of 7.37%[21][24] - **120-day Volatility**: Weekly return of 0.50%, monthly return of -4.08%, and YTD return of 11.22%[21][24]

Quantitative Models and Construction 1. Model Name: Barra Style Factors - **Model Construction Idea**: The Barra style factors are designed to capture various market characteristics such as valuation, momentum, volatility, and growth, among others, to explain stock returns[14][15] - **Model Construction Process**: - The factors are calculated based on specific financial and market metrics. For example: - **Beta**: Historical beta - **Size**: Natural logarithm of total market capitalization - **Momentum**: Weighted average of historical excess return series - **Volatility**: Weighted average of historical residual return volatility - **Valuation**: Inverse of price-to-book ratio - **Liquidity**: Weighted average of turnover ratios (monthly, quarterly, yearly) - **Profitability**: Weighted average of various profitability metrics such as analyst forecasted earnings-to-price ratio, inverse of price-to-cash flow ratio, and inverse of trailing twelve-month price-to-earnings ratio - **Growth**: Weighted average of earnings growth rate and revenue growth rate - **Leverage**: Weighted average of market leverage, book leverage, and debt-to-asset ratio[15] - **Model Evaluation**: The model is widely used in the industry to capture systematic risk factors and explain stock returns. It is considered robust and comprehensive in its approach to factor construction[14][15] 2. Model Name: GRU (Generalized Risk Utility) Model - **Model Construction Idea**: GRU models are used to capture complex relationships in stock returns by leveraging advanced statistical and machine learning techniques. They are designed to identify patterns in historical data and predict future performance[4][6][8] - **Model Construction Process**: - GRU models are trained on historical data to identify patterns in stock returns - The models are applied to different stock pools (e.g., CSI 300, CSI 500, CSI 1000) to evaluate their performance - Specific GRU models include `barra1d`, `barra5d`, `open1d`, and `close1d`, which differ in their time horizons and data inputs[4][6][8] - **Model Evaluation**: GRU models show mixed performance, with some models like `barra5d` and `close1d` performing strongly, while others like `barra1d` exhibit significant drawdowns in certain periods[4][6][8] --- Model Backtesting Results 1. Barra Style Factors - **Momentum**: Weekly return 3.49%, monthly return -6.50%, YTD return -14.88%[17] - **Beta**: Weekly return 2.21%, monthly return -7.75%, YTD return 28.44%[17] - **Volatility**: Weekly return 1.90%, monthly return -3.76%, YTD return 6.09%[17] - **Liquidity**: Weekly return 1.67%, monthly return 46.39%, YTD return 8.77%[17] - **Size**: Weekly return 0.45%, monthly return -6.89%, YTD return -39.47%[17] - **Non-linear Size**: Weekly return 0.28%, monthly return -6.47%, YTD return -34.37%[17] - **Growth**: Weekly return 0.22%, monthly return 2.03%, YTD return 0.89%[17] - **Profitability**: Weekly return 1.43%, monthly return 3.55%, YTD return 14.39%[17] - **Leverage**: Weekly return 2.13%, monthly return 4.08%, YTD return 16.59%[17] - **Valuation**: Weekly return 3.52%, monthly return 6.78%, YTD return 4.37%[17] 2. GRU Models - **barra1d**: Weekly return -0.34%, monthly return -0.65%, YTD return 4.71%[33][34] - **barra5d**: Weekly return 1.44%, monthly return 5.42%, YTD return 7.34%[33][34] - **open1d**: Weekly return 0.32%, monthly return 1.81%, YTD return 6.02%[33][34] - **close1d**: Weekly return 1.41%, monthly return 4.17%, YTD return 4.33%[33][34] - **Multi-factor Combination**: Weekly return 0.57%, monthly return 2.54%, YTD return 0.89%[33][34] --- Quantitative Factors and Construction 1. Factor Name: Fundamental Factors - **Factor Construction Idea**: Fundamental factors are derived from financial metrics to capture the underlying financial health and performance of companies[4][6][7] - **Factor Construction Process**: - Metrics such as return on assets (ROA), return on equity (ROE), and revenue growth are calculated using trailing twelve-month (TTM) data - Factors are industry-neutralized before testing[19] - **Factor Evaluation**: Fundamental factors show mixed performance, with some factors like "growth" and "profitability" performing well, while others like "static financial factors" exhibit negative returns in certain periods[4][6][7] 2. Factor Name: Technical Factors - **Factor Construction Idea**: Technical factors are based on price and volume data to capture market trends and investor behavior[4][6][7] - **Factor Construction Process**: - Metrics such as momentum, volatility, and turnover are calculated over different time horizons (e.g., 20-day, 60-day, 120-day) - Factors are industry-neutralized before testing[19] - **Factor Evaluation**: Technical factors generally show positive returns for momentum-based factors, while volatility-based factors often exhibit negative returns[4][6][7] --- Factor Backtesting Results 1. Fundamental Factors (CSI 300) - **ROA Growth**: Weekly return 0.38%, monthly return 2.38%, YTD return 26.31%[23] - **Net Profit Surprise Growth**: Weekly return 1.10%, monthly return 2.62%, YTD return 42.59%[23] - **ROC Surprise Growth**: Weekly return 2.23%, monthly return 2.23%, YTD return 35.35%[23] 2. Technical Factors (CSI 500) - **20-day Momentum**: Weekly return 5.99%, monthly return 1.74%, YTD return 3.65%[26] - **120-day Momentum**: Weekly return 1.76%, monthly return 4.01%, YTD return 3.55%[26] - **20-day Volatility**: Weekly return -1.15%, monthly return -4.31%, YTD return 25.86%[26]

- The report tracks the performance of style factors, including liquidity, volatility, and nonlinear market capitalization, which showed strong long positions, while valuation, profitability, and leverage factors exhibited strong short positions [2][16] - Barra style factors are constructed using various financial and technical metrics, such as historical beta, logarithm of total market capitalization, historical excess return momentum, and volatility calculated as a weighted combination of historical excess return volatility, cumulative excess return deviation, and residual return volatility [14][15] - Liquidity factor is calculated as a weighted combination of monthly turnover rate (35%), quarterly turnover rate (35%), and annual turnover rate (30%) [15] - Profitability factor is constructed using a weighted combination of analyst forecast earnings-to-price ratio (68%), inverse cash flow ratio (21%), inverse PE ratio (11%), forecast long-term earnings growth rate (18%), and forecast short-term earnings growth rate (11%) [15] - Growth factor is calculated using a weighted combination of earnings growth rate (24%) and revenue growth rate (47%) [15] - Leverage factor is constructed using market leverage ratio (38%), book leverage (35%), and asset-liability ratio (27%) [15] - GRU models, including open1d, close1d, barra1d, and barra5d, are tracked for their multi-factor performance across different stock pools, showing varied results in terms of long-short returns [3][4][5][6] - GRU models demonstrated strong performance in certain configurations, such as close1d and barra5d, while open1d and barra1d showed weaker returns in specific periods [31][33] - Multi-factor portfolios underperformed this week, with relative excess returns against the CSI 1000 index showing a decline of 0.95% [33][34] - Barra5d model exhibited strong year-to-date performance, achieving an excess return of 5.81% against the CSI 1000 index [33][34] - Technical factors, including short-term and long-term momentum and volatility metrics, showed mixed results across different stock pools, with short-term metrics generally outperforming [19][21][24][26] - Basic financial factors, such as static financial metrics and growth-related metrics, generally showed negative long-short returns, with low-growth stocks outperforming [19][21][24][26] - GRU models' long-short returns varied across stock pools, with close1d and barra5d models showing strong positive returns, while open1d and barra1d models experienced slight pullbacks [31][33] - The liquidity factor achieved a weekly return of 1.39%, while the volatility factor returned 0.92% over the same period [17] - Profitability factor showed a weekly return of -1.31%, and valuation factor returned -1.53% [17] - Growth factor achieved a weekly return of 0.21%, while leverage factor returned -0.83% [17] - GRU models' weekly returns included -0.82% for open1d, 2.88% for close1d, -0.45% for barra1d, and 1.23% for barra5d [31] - Multi-factor portfolio weekly return was -0.95% relative to the CSI 1000 index [34]

- **Barra style factors**: The report tracks several style factors including Beta, Market Cap, Momentum, Volatility, Non-linear Market Cap, Valuation, Liquidity, Profitability, Growth, and Leverage. These factors are constructed using historical data and financial metrics such as turnover rates, earnings growth rates, and market leverage ratios. For example, the Beta factor represents historical beta, while the Valuation factor is calculated as the inverse of the price-to-book ratio. The formulas for constructing these factors include weighted combinations of metrics like turnover rates and earnings ratios [14][15][16] - **Factor performance tracking**: The report evaluates the recent performance of style factors across the market. Beta, Liquidity, and Momentum factors showed strong long positions, while Market Cap, Non-linear Market Cap, and Valuation factors performed better in short positions. The tracking methodology involves selecting stocks from the Wind All A pool, excluding ST stocks, suspended stocks, and newly listed stocks under 120 days. Long positions are taken in the top 10% of stocks with the highest factor values, and short positions in the bottom 10%, with equal weight allocation [16][19][20] - **Factor backtesting results**: The report provides detailed backtesting results for style factors. For example, Beta achieved a weekly return of 4.58%, while Market Cap showed a negative weekly return of -3.55%. Other factors like Momentum and Liquidity also demonstrated varied performance across different time horizons, such as one week, one month, and year-to-date. The report highlights the annualized returns for three-year and five-year periods for each factor [17][18][19] - **GRU factor performance**: GRU factors showed weaker performance overall, with only the barra1d model achieving positive returns. Other GRU models experienced drawdowns in their long-short portfolios. This indicates potential challenges in the effectiveness of GRU factors under current market conditions [20][25][29] - **Technical factors**: Technical factors such as 20-day Momentum, 60-day Momentum, and various volatility measures (e.g., 120-day Volatility) were tracked. These factors generally showed positive returns in long positions, particularly in high-volatility and high-momentum stocks. For example, 120-day Volatility achieved a weekly return of 5.92% in the CSI 300 stock pool [24][27][31] - **Fundamental factors**: Fundamental factors like ROA growth, ROC growth, and Net Profit growth were analyzed. In the CSI 300 stock pool, Net Profit growth achieved a weekly return of 2.51%, while ROA growth showed a return of 1.19%. These factors generally favored stocks with stable and strong growth metrics [23][25][30] - **Multi-factor portfolio performance**: The report evaluates the performance of multi-factor portfolios. The barra5d model outperformed the CSI 1000 index by 0.27% this week and achieved a year-to-date excess return of 5.91%. Other models showed mixed results, with some experiencing slight drawdowns. The multi-factor portfolio achieved a weekly excess return of 0.04% relative to the CSI 1000 index [8][33][34]

Quantitative Models and Factor Analysis Quantitative Models and Construction - **Model Name**: GRU-based Models - **Construction Idea**: GRU (Gated Recurrent Unit) models are used to capture sequential patterns in financial data, aiming to predict stock movements based on historical trends and other input features [3][4][5] - **Construction Process**: GRU models are trained on historical data to optimize their predictive capabilities. Specific variations of GRU models include `barra1d`, `barra5d`, `open1d`, and `close1d`, which differ in their input features and time horizons [3][4][5] - **Evaluation**: GRU models show mixed performance, with `barra1d` consistently achieving positive returns, while other variations like `close1d` and `barra5d` experience significant drawdowns [3][4][5] Model Backtesting Results - **GRU Models**: - `barra1d`: Weekly excess return of 0.14%, monthly return of 1.20%, and YTD return of 4.77% [32][33] - `barra5d`: Weekly excess return of -0.59%, monthly return of -2.84%, and YTD return of 5.03% [32][33] - `open1d`: Weekly excess return of 0.22%, monthly return of -1.23%, and YTD return of 5.45% [32][33] - `close1d`: Weekly excess return of -0.20%, monthly return of -2.64%, and YTD return of 2.92% [32][33] --- Quantitative Factors and Construction - **Factor Name**: Style Factors (Barra) - **Construction Idea**: Style factors are designed to capture systematic risks and returns associated with specific stock characteristics, such as size, momentum, and valuation [14][15] - **Construction Process**: - **Beta**: Historical beta of the stock - **Size**: Natural logarithm of total market capitalization - **Momentum**: Mean of historical excess returns - **Volatility**: Weighted combination of historical excess return volatility, cumulative excess return deviation, and residual return volatility - **Valuation**: Inverse of price-to-book ratio - **Liquidity**: Weighted turnover rates over monthly, quarterly, and yearly periods - **Profitability**: Weighted combination of analyst-predicted earnings yield, cash flow yield, and other profitability metrics - **Growth**: Weighted combination of earnings and revenue growth rates - **Leverage**: Weighted combination of market leverage, book leverage, and debt-to-asset ratio [15] - **Evaluation**: Style factors exhibit varying performance, with size, non-linear size, and liquidity factors showing strong long positions, while valuation and growth factors perform better in short positions [16][17] - **Factor Name**: Fundamental Factors - **Construction Idea**: Fundamental factors are derived from financial statements and aim to capture the financial health and growth potential of companies [17][18][20] - **Construction Process**: - **ROA Growth**: Growth in return on assets - **ROC Growth**: Growth in return on capital - **Net Profit Growth**: Growth in net profit - **Sales-to-Price Ratio**: Inverse of price-to-sales ratio - **Operating Profit Growth**: Growth in operating profit [21][25][27] - **Evaluation**: Fundamental factors like ROA and ROC growth show positive returns, while static financial metrics like sales-to-price ratio exhibit mixed results [21][25][27] - **Factor Name**: Technical Factors - **Construction Idea**: Technical factors are based on price and volume data, aiming to capture momentum and volatility patterns [18][20][24] - **Construction Process**: - **Momentum**: Calculated over 20, 60, and 120-day periods - **Volatility**: Measured over similar time horizons - **Median Deviation**: Deviation of stock prices from the median [25][27][30] - **Evaluation**: High-momentum stocks generally outperform, while long-term volatility factors show weaker performance [25][27][30] --- Factor Backtesting Results - **Style Factors**: - Size: Weekly return of 0.22%, monthly return of 1.20%, and YTD return of 4.77% [16][17] - Valuation: Weekly return of -0.20%, monthly return of -2.64%, and YTD return of 2.92% [16][17] - **Fundamental Factors**: - ROA Growth: Weekly return of 1.31%, monthly return of 12.03%, and YTD return of 33.49% [21][25] - ROC Growth: Weekly return of 1.74%, monthly return of 4.75%, and YTD return of 10.89% [21][25] - **Technical Factors**: - 20-day Momentum: Weekly return of 3.25%, monthly return of 12.92%, and YTD return of 2.35% [25][27] - 60-day Volatility: Weekly return of 3.65%, monthly return of 16.15%, and YTD return of 28.43% [25][27]

Quantitative Models and Construction 1. Model Name: barra1d - **Model Construction Idea**: This model is part of the GRU factor family and is designed to capture short-term market dynamics through daily data inputs[4][6][8] - **Model Construction Process**: The barra1d model uses daily market data to calculate factor exposures and returns. It applies industry-neutralization and standardization processes to ensure comparability across stocks. The model is rebalanced monthly, selecting the top 10% of stocks with the highest factor scores for long positions and the bottom 10% for short positions, with equal weighting[17][28][29] - **Model Evaluation**: The barra1d model demonstrated strong performance in multiple stock pools, showing resilience in volatile market conditions[4][6][8] 2. Model Name: barra5d - **Model Construction Idea**: This model extends the barra1d framework to a five-day horizon, aiming to capture slightly longer-term market trends[4][6][8] - **Model Construction Process**: Similar to barra1d, the barra5d model uses five-day aggregated data for factor calculation. It follows the same industry-neutralization, standardization, and rebalancing processes as barra1d[17][28][29] - **Model Evaluation**: The barra5d model experienced significant drawdowns in recent periods, indicating sensitivity to market reversals[4][6][8] 3. Model Name: open1d - **Model Construction Idea**: This model focuses on open price data to identify short-term trading opportunities[4][6][8] - **Model Construction Process**: The open1d model calculates factor exposures based on daily opening prices. It applies the same industry-neutralization and rebalancing methodology as other GRU models[17][28][29] - **Model Evaluation**: The open1d model showed moderate performance, with some drawdowns in recent periods[4][6][8] 4. Model Name: close1d - **Model Construction Idea**: This model emphasizes closing price data to capture end-of-day market sentiment[4][6][8] - **Model Construction Process**: The close1d model uses daily closing prices for factor calculation. It follows the same construction and rebalancing methodology as other GRU models[17][28][29] - **Model Evaluation**: The close1d model demonstrated stable performance, with positive returns in certain stock pools[4][6][8] --- Model Backtesting Results 1. barra1d Model - Weekly Excess Return: +0.57%[29][30] - Monthly Excess Return: +0.75%[29][30] - Year-to-Date Excess Return: +4.38%[29][30] 2. barra5d Model - Weekly Excess Return: -2.17%[29][30] - Monthly Excess Return: -3.76%[29][30] - Year-to-Date Excess Return: +4.13%[29][30] 3. open1d Model - Weekly Excess Return: -0.97%[29][30] - Monthly Excess Return: -2.85%[29][30] - Year-to-Date Excess Return: +4.20%[29][30] 4. close1d Model - Weekly Excess Return: -1.68%[29][30] - Monthly Excess Return: -4.50%[29][30] - Year-to-Date Excess Return: +1.90%[29][30] --- Quantitative Factors and Construction 1. Factor Name: Beta - **Factor Construction Idea**: Measures historical market sensitivity of a stock[15] - **Factor Construction Process**: Calculated as the regression coefficient of a stock's returns against market returns over a specified period[15] 2. Factor Name: Size - **Factor Construction Idea**: Captures the size effect, where smaller firms tend to outperform larger ones[15] - **Factor Construction Process**: Defined as the natural logarithm of total market capitalization[15] 3. Factor Name: Momentum - **Factor Construction Idea**: Identifies stocks with strong recent performance[15] - **Factor Construction Process**: Combines historical excess return mean, volatility, and cumulative deviation into a weighted formula: $ Momentum = 0.74 * \text{Volatility} + 0.16 * \text{Cumulative Deviation} + 0.10 * \text{Residual Volatility} $[15] 4. Factor Name: Volatility - **Factor Construction Idea**: Measures the risk or variability in stock returns[15] - **Factor Construction Process**: Weighted combination of historical residual volatility and other measures[15] 5. Factor Name: Valuation - **Factor Construction Idea**: Captures the value effect, where undervalued stocks tend to outperform[15] - **Factor Construction Process**: Defined as the inverse of the price-to-book ratio[15] 6. Factor Name: Liquidity - **Factor Construction Idea**: Measures the ease of trading a stock[15] - **Factor Construction Process**: Weighted combination of turnover rates over monthly, quarterly, and yearly horizons: $ Liquidity = 0.35 * \text{Monthly Turnover} + 0.35 * \text{Quarterly Turnover} + 0.30 * \text{Yearly Turnover} $[15] 7. Factor Name: Profitability - **Factor Construction Idea**: Identifies stocks with strong earnings performance[15] - **Factor Construction Process**: Weighted combination of various profitability metrics, including analyst forecasts and financial ratios[15] 8. Factor Name: Growth - **Factor Construction Idea**: Captures the growth potential of a stock[15] - **Factor Construction Process**: Weighted combination of earnings and revenue growth rates[15] --- Factor Backtesting Results 1. Beta Factor - Weekly Return: +0.14%[21] - Monthly Return: +1.65%[21] - Year-to-Date Return: +5.29%[21] 2. Size Factor - Weekly Return: +0.36%[21] - Monthly Return: +1.00%[21] - Year-to-Date Return: +6.37%[21] 3. Momentum Factor - Weekly Return: +2.21%[24] - Monthly Return: +8.80%[24] - Year-to-Date Return: +23.30%[24] 4. Volatility Factor - Weekly Return: +2.82%[24] - Monthly Return: +12.29%[24] - Year-to-Date Return: +25.25%[24] 5. Valuation Factor - Weekly Return: +1.47%[21] - Monthly Return: +2.30%[21] - Year-to-Date Return: -2.26%[21] 6. Liquidity Factor - Weekly Return: +1.80%[21] - Monthly Return: +5.91%[21] - Year-to-Date Return: +19.70%[21] 7. Profitability Factor - Weekly Return: +4.57%[21] - Monthly Return: +7.53%[21] - Year-to-Date Return: +27.56%[21] 8. Growth Factor - Weekly Return: +2.76%[24] - Monthly Return: +6.51%[24] - Year-to-Date Return: +14.51%[24]

Quantitative Models and Construction 1. Model Name: GRU Model - **Model Construction Idea**: The GRU model is used to predict stock returns based on historical data and incorporates various factors to optimize portfolio performance [3][4][5] - **Model Construction Process**: - The GRU model is trained on historical data to capture temporal dependencies in stock returns - It uses multiple input features, including technical and fundamental factors, to predict future returns - The model is applied to different stock pools (e.g., CSI 300, CSI 500, CSI 1000) to evaluate its performance [5][6][7] - **Model Evaluation**: The GRU model demonstrates strong performance in most stock pools, with positive long-short returns across various factors. However, certain sub-models (e.g., `barra5d`) show occasional underperformance [5][6][7] 2. Model Name: Open1d and Close1d Models - **Model Construction Idea**: These models focus on short-term price movements and are designed to capture daily return patterns [8][31] - **Model Construction Process**: - Open1d and Close1d models are trained on daily open and close price data, respectively - They are evaluated based on their ability to generate excess returns relative to the CSI 1000 index [8][31] - **Model Evaluation**: These models show mixed performance, with occasional drawdowns relative to the benchmark index [8][31] 3. Model Name: Barra1d and Barra5d Models - **Model Construction Idea**: These models are based on the Barra factor framework and aim to capture short-term and medium-term return patterns [8][31] - **Model Construction Process**: - Barra1d focuses on daily factor returns, while Barra5d aggregates returns over a 5-day horizon - Both models are tested for their ability to generate excess returns relative to the CSI 1000 index [8][31] - **Model Evaluation**: Barra5d demonstrates strong year-to-date performance, significantly outperforming the benchmark, while Barra1d shows consistent but less pronounced gains [8][31] --- Model Backtest Results 1. GRU Model - **Excess Return**: Positive across most stock pools, with occasional underperformance in specific sub-models like `barra5d` [5][6][7] 2. Open1d Model - **Weekly Excess Return**: -0.01% - **Year-to-Date Excess Return**: 5.23% [32] 3. Close1d Model - **Weekly Excess Return**: -0.38% - **Year-to-Date Excess Return**: 3.64% [32] 4. Barra1d Model - **Weekly Excess Return**: 0.65% - **Year-to-Date Excess Return**: 3.80% [32] 5. Barra5d Model - **Weekly Excess Return**: 0.02% - **Year-to-Date Excess Return**: 6.44% [32] --- Quantitative Factors and Construction 1. Factor Name: Beta - **Factor Construction Idea**: Measures historical beta to capture market sensitivity [15] - **Factor Construction Process**: Historical beta is calculated based on the covariance of stock returns with market returns [15] 2. Factor Name: Momentum - **Factor Construction Idea**: Captures historical excess return trends [15] - **Factor Construction Process**: - Momentum = 0.74 * Historical Excess Return Volatility + 0.16 * Cumulative Excess Return Deviation + 0.1 * Historical Residual Return Volatility [15] 3. Factor Name: Volatility - **Factor Construction Idea**: Measures stock price fluctuations to identify high-volatility stocks [15] - **Factor Construction Process**: - Volatility = Weighted combination of historical residual return volatility and other metrics [15] 4. Factor Name: Growth - **Factor Construction Idea**: Focuses on earnings and revenue growth rates [15] - **Factor Construction Process**: - Growth = 0.24 * Earnings Growth Rate + 0.47 * Revenue Growth Rate [15] 5. Factor Name: Liquidity - **Factor Construction Idea**: Measures stock turnover to identify liquid stocks [15] - **Factor Construction Process**: - Liquidity = 0.35 * Monthly Turnover + 0.35 * Quarterly Turnover + 0.3 * Annual Turnover [15] --- Factor Backtest Results 1. Beta Factor - **Weekly Long-Short Return**: Positive [16][18] 2. Momentum Factor - **Weekly Long-Short Return**: Negative [16][18] 3. Volatility Factor - **Weekly Long-Short Return**: Positive [16][18] 4. Growth Factor - **Weekly Long-Short Return**: Positive [16][18] 5. Liquidity Factor - **Weekly Long-Short Return**: Positive [16][18]

Quantitative Models and Construction 1. Model Name: GRU Model - **Model Construction Idea**: The GRU model is used to capture temporal dependencies in financial data, leveraging its recurrent structure to predict stock movements and generate long-short signals[4][5][6] - **Model Construction Process**: - Input data includes historical stock prices, technical indicators, and fundamental factors - The GRU network processes sequential data to learn patterns over time - Outputs are used to construct long-short portfolios based on predicted returns[4][5][6] - **Model Evaluation**: The GRU model demonstrates strong performance in certain market conditions, though its results vary across different stock pools[4][5][6] 2. Model Name: Barra Models (barra1d, barra5d) - **Model Construction Idea**: Barra models are factor-based models designed to decompose stock returns into systematic and idiosyncratic components, enabling factor-based portfolio construction[4][5][6] - **Model Construction Process**: - Factors such as size, value, momentum, and volatility are calculated for each stock - Stocks are ranked based on factor scores, and portfolios are constructed by going long the top 10% and short the bottom 10% of stocks based on factor rankings - barra1d uses daily data, while barra5d aggregates data over a 5-day window[4][5][6] - **Model Evaluation**: barra1d shows consistent strong performance, while barra5d experiences significant drawdowns in certain periods[4][5][6] --- Backtesting Results of Models GRU Model - **Open1d**: Weekly excess return: -1.80%, Monthly: -1.96%, YTD: 5.24%[33] - **Close1d**: Weekly excess return: -2.40%, Monthly: -3.10%, YTD: 4.04%[33] Barra Models - **Barra1d**: Weekly excess return: -0.63%, Monthly: -0.34%, YTD: 3.13%[33] - **Barra5d**: Weekly excess return: -1.80%, Monthly: -2.08%, YTD: 6.42%[33] --- Quantitative Factors and Construction 1. Factor Name: Beta - **Factor Construction Idea**: Measures the sensitivity of a stock's returns to market movements[15] - **Factor Construction Process**: Calculated as the historical beta of the stock relative to the market index[15] 2. Factor Name: Size - **Factor Construction Idea**: Captures the size effect, where smaller firms tend to outperform larger firms[15] - **Factor Construction Process**: Natural logarithm of total market capitalization[15] 3. Factor Name: Momentum - **Factor Construction Idea**: Stocks with strong past performance tend to continue performing well in the short term[15] - **Factor Construction Process**: - Weighted combination of historical excess return volatility (0.74), cumulative excess return deviation (0.16), and residual return volatility (0.10)[15] 4. Factor Name: Volatility - **Factor Construction Idea**: Measures the risk or variability in stock returns[15] - **Factor Construction Process**: Weighted combination of historical residual return volatility and other metrics[15] 5. Factor Name: Valuation - **Factor Construction Idea**: Identifies undervalued stocks based on fundamental metrics[15] - **Factor Construction Process**: Inverse of price-to-book ratio[15] 6. Factor Name: Liquidity - **Factor Construction Idea**: Measures the ease of trading a stock[15] - **Factor Construction Process**: Weighted combination of monthly turnover (0.35), quarterly turnover (0.35), and annual turnover (0.30)[15] 7. Factor Name: Profitability - **Factor Construction Idea**: Captures the financial health and earnings quality of a firm[15] - **Factor Construction Process**: Weighted combination of analyst-predicted earnings yield, cash flow yield, and other profitability metrics[15] 8. Factor Name: Growth - **Factor Construction Idea**: Identifies firms with strong earnings and revenue growth[15] - **Factor Construction Process**: Weighted combination of earnings growth rate (0.24) and revenue growth rate (0.47)[15] 9. Factor Name: Leverage - **Factor Construction Idea**: Measures the financial risk associated with a firm's debt levels[15] - **Factor Construction Process**: Weighted combination of market leverage (0.38), book leverage (0.35), and debt-to-asset ratio (0.27)[15] --- Backtesting Results of Factors Fundamental Factors - **Growth**: Weekly excess return: 2.41%, Monthly: -2.18%, YTD: 3.20%[28] - **Profitability**: Weekly excess return: 0.22%, Monthly: 40.98%, YTD: 6.12%[28] Technical Factors - **Momentum (20-day)**: Weekly excess return: 1.72%, Monthly: 4.23%, YTD: -5.29%[30] - **Volatility (120-day)**: Weekly excess return: 4.85%, Monthly: 8.64%, YTD: -14.60%[30]