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: CSI 500 Enhanced Portfolio - **Model Construction Idea**: The model aims to generate excess returns relative to the CSI 500 index by leveraging a quantitative strategy based on factor models and portfolio optimization techniques [45] - **Model Construction Process**: - The portfolio is constructed using a strategy model that selects stocks based on specific quantitative factors [45] - The portfolio weights are optimized to maximize the expected return while controlling for risk and tracking error relative to the CSI 500 index [45] - The model's performance is evaluated on a weekly basis, and adjustments are made to the portfolio as needed [45] - **Model Evaluation**: The model has demonstrated significant excess returns over the CSI 500 index since 2020, though it experienced underperformance in the most recent week [45] 2. Model Name: CSI 300 Enhanced Portfolio - **Model Construction Idea**: Similar to the CSI 500 Enhanced Portfolio, this model seeks to outperform the CSI 300 index using quantitative factor-based strategies and portfolio optimization [51] - **Model Construction Process**: - Stocks are selected based on quantitative factors, and portfolio weights are optimized to achieve excess returns while managing risk and tracking error relative to the CSI 300 index [51] - The portfolio is reviewed and adjusted periodically to align with the strategy model's recommendations [51] - **Model Evaluation**: The model has achieved consistent excess returns over the CSI 300 index since 2020, with a slight outperformance in the most recent week [51] --- Model Backtesting Results CSI 500 Enhanced Portfolio - Weekly return: 1.82% - Underperformance relative to the benchmark: -1.56% - Cumulative excess return since 2020: 49.43% - Maximum drawdown: -4.99% [45] CSI 300 Enhanced Portfolio - Weekly return: 1.40% - Outperformance relative to the benchmark: 0.02% - Cumulative excess return since 2020: 39.41% - Maximum drawdown: -5.86% [51] --- Quantitative Factors and Construction 1. Factor Name: Beta - **Factor Construction Idea**: Measures the sensitivity of a stock's returns to market movements, capturing the systematic risk of the stock [55] - **Factor Construction Process**: - Beta is calculated using regression analysis of a stock's returns against the market index returns over a specified period [55] - The formula is: $ \beta = \frac{\text{Cov}(R_i, R_m)}{\text{Var}(R_m)} $ where $R_i$ is the stock return, $R_m$ is the market return, Cov is covariance, and Var is variance [55] - **Factor Evaluation**: High Beta stocks have recently outperformed, reflecting a market preference for higher systematic risk [56] 2. Factor Name: Residual Volatility (RESVOL) - **Factor Construction Idea**: Captures the idiosyncratic risk of a stock, representing the volatility of its returns unexplained by market movements [55] - **Factor Construction Process**: - Residual volatility is derived from the standard deviation of the residuals in a regression of stock returns on market returns [55] - The formula is: $ \text{RESVOL} = \sqrt{\frac{\sum (R_i - \alpha - \beta R_m)^2}{n-2}} $ where $R_i$ is the stock return, $R_m$ is the market return, $\alpha$ is the intercept, $\beta$ is the slope, and $n$ is the number of observations [55] - **Factor Evaluation**: Residual volatility has shown a significant negative excess return in the recent period, indicating underperformance of high idiosyncratic risk stocks [56] 3. Factor Name: Nonlinear Size (NLSIZE) - **Factor Construction Idea**: Captures the nonlinear relationship between stock size and returns, complementing the traditional size factor [55] - **Factor Construction Process**: - Nonlinear size is calculated as the square of the logarithm of market capitalization: $ \text{NLSIZE} = (\log(\text{Market Cap}))^2 $ [55] - **Factor Evaluation**: Nonlinear size has underperformed recently, reflecting a lack of market preference for mid-sized stocks [56] --- Factor Backtesting Results Beta Factor - Weekly pure factor return: Positive [56] Residual Volatility Factor - Weekly pure factor return: Negative [56] Nonlinear Size Factor - Weekly pure factor return: Negative [56]

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]

Core Insights - Style factors serve as a stable foundation for stock portfolios, with changes in factor correlations expected to create tactical Alpha opportunities [2][19] - Chinese stock factors exhibit significant long-term risk-return characteristics, with style factors showing statistically significant risk-adjusted returns exceeding market factors [2][4] Long-term Performance of Pure Factor Portfolios - Over the past 20 years (2005-2024), the annualized return for the market factor is 4.2% with a volatility of 26.9%, resulting in a risk-adjusted return of only 0.16, which is statistically insignificant [2] - Among 14 style factors analyzed, nine demonstrated statistically significant risk-adjusted returns, with a threshold of 0.44 for significance [2] - Notable annualized returns for specific factors include: - Momentum: 4.9% - Earnings: 2.7% - Valuation: 2.1% - Growth: 1.7% - Beta: 6.6% [2] Recent Performance of Style Factors - The returns of style factors vary based on market mechanisms, particularly the beta factor, which shows a return of 1.5% in the top quintile of market returns and drops to -0.1% in the bottom quintile [6] - Earnings yield, beta, momentum, and profit factors exhibit significant volatility depending on market performance [6] Changes in Factor Correlations - The correlation between long-term value and beta has shifted from near zero to a historical low of -0.65, indicating a growing presence of stocks with both value and defensive characteristics [13] - Momentum's correlation with beta also decreased to -0.70 but has since rebounded to -0.40, while the correlation between value and momentum has returned to near zero from a historical high of 0.78 [13] Tactical Alpha Opportunities - The dispersion of style factor returns is currently above average, with a 12-week moving average dispersion of 0.45%, close to the 75th percentile of the past 20 years [19] - High dispersion indicates inconsistent factor performance, potentially allowing skilled active managers to achieve Alpha by deviating from benchmark indices [19]

Quantitative Models and Construction Methods Model Name: Black Swan Index - **Construction Idea**: Measure the extremity of market transactions based on the deviation of style factor returns[24][25] - **Construction Process**: 1. Calculate the daily return deviation of style factors: $$ \sigma_{s,t}=\frac{\bar{r}_{s,t}-\bar{r}_{s}}{\sigma_{s}} $$ where $\bar{r}_{s,t}$ is the daily return of style factor $s$ on day $t$, $\bar{r}_{s}$ is the average daily return of style factor $s$ over the entire sample period, and $\sigma_{s}$ is the standard deviation of daily returns of style factor $s$ over the entire sample period[25] 2. Calculate the Black Swan Index: $$ BlackSwan_{t}=\frac{1}{N}\times\sum_{s\in S}\left|\sigma_{s,t}\right| $$ where $BlackSwan_{t}$ is the Black Swan Index on day $t$, $S$ is the set of all style factors, and $N$ is the number of style factors[25] - **Evaluation**: The Black Swan Index effectively captures the extremity of market transactions, indicating higher probabilities of extreme tail risks[24][25] Model Name: Heuristic Style Classification for Cognitive Risk Control - **Construction Idea**: Address the discrepancy between individual and collective cognition in style classification to control cognitive risk[80][81] - **Construction Process**: 1. Calculate the value and growth factors for each stock based on predefined metrics[85] 2. Construct value and growth portfolios by selecting the top 10% and bottom 10% stocks based on factor scores[82] 3. Perform time-series regression to classify stocks into value, growth, or balanced styles: $$ r_{t,t}\sim\beta_{\mathit{Value}}\cdot r_{\mathit{Value},t}+\beta_{\mathit{Growth}}\cdot r_{\mathit{Growth},t}+\varepsilon_{t} $$ subject to $0\leq\beta_{\mathit{Value}}\leq1$, $0\leq\beta_{\mathit{Growth}}\leq1$, and $\beta_{\mathit{Value}}+\beta_{\mathit{Growth}}=1$[97] 4. Use weighted least squares (WLS) to estimate regression coefficients based on the most differentiated trading days[98] - **Evaluation**: The heuristic style classification method captures market consensus more accurately than traditional factor scoring methods, reducing cognitive risk[80][81] Model Name: Louvain Community Detection for Hidden Risk Control - **Construction Idea**: Cluster stocks based on excess return correlations to identify hidden risks[116][117] - **Construction Process**: 1. Calculate weighted correlation of excess returns between stocks: $$ Corr_{w}(X,Y)=\frac{Cov_{w}(X,Y)}{\sigma_{w,X}\cdot\sigma_{w,Y}}=\frac{\sum_{i=1}^{n}w_{i}(x_{i}-\overline{X_{w}})(y_{i}-\overline{Y_{w}})}{\sqrt{\sum_{i=1}^{n}w_{i}(x_{i}-\overline{X_{w}})^{2}}\cdot\sqrt{\sum_{i=1}^{n}w_{i}(y_{i}-\overline{Y_{w}})^{2}}} $$ where $w_{i}$ is the weight for day $i$, reflecting market volatility[118] 2. Use Louvain algorithm to cluster stocks based on weighted correlation matrix[117] 3. Ensure clusters have at least 20 stocks and remove clusters with fewer stocks[121] - **Evaluation**: The Louvain community detection method effectively identifies hidden risks by clustering stocks with similar return patterns, which traditional risk models may overlook[116][117] Model Name: Dynamic Style Factor Control - **Construction Idea**: Control style factors dynamically based on their volatility clustering effect[128][129] - **Construction Process**: 1. Identify style factors with high volatility or significant volatility increase: $$ \text{High volatility: Rolling 3-month volatility in top 3} $$ $$ \text{Volatility increase: Rolling 3-month volatility > historical mean + 1 standard deviation} $$ 2. Set the exposure of these style factors to zero in the portfolio[136] - **Evaluation**: Dynamic style factor control captures major market risks without significantly affecting portfolio returns, leveraging the predictability of volatility clustering[128][129] Model Name: Adaptive Stock Deviation Control under Target Tracking Error - **Construction Idea**: Adjust stock deviation based on tracking error to control portfolio risk[146][147] - **Construction Process**: 1. Calculate rolling 3-month tracking error for different stock deviation levels[153] 2. Set the maximum stock deviation that keeps tracking error within the target range[153] - **Evaluation**: Adaptive stock deviation control effectively reduces tracking error during high market volatility, maintaining portfolio stability[146][147] Model Backtest Results Traditional CSI 500 Enhanced Index - **Annualized Excess Return**: 18.77%[5][162] - **Maximum Drawdown**: 9.68%[5][162] - **Information Ratio (IR)**: 3.56[5][162] - **Return-to-Drawdown Ratio**: 1.94[5][162] - **Annualized Tracking Error**: 4.88%[5][162] CSI 500 Enhanced Index with Full-Process Risk Control - **Annualized Excess Return**: 16.51%[5][169] - **Maximum Drawdown**: 4.90%[5][169] - **Information Ratio (IR)**: 3.94[5][169] - **Return-to-Drawdown Ratio**: 3.37[5][169] - **Annualized Tracking Error**: 3.98%[5][169]

Quantitative Models and Construction Methods - **Model Name**: GRU **Model Construction Idea**: GRU is used for industry rotation and stock selection based on historical data and market trends[3][5][7] **Model Construction Process**: GRU utilizes gated recurrent units to process sequential data, capturing temporal dependencies in stock price movements and industry performance. It incorporates multiple factors such as momentum, volatility, and valuation metrics to predict future trends[3][5][7] **Model Evaluation**: GRU demonstrates strong performance in multi-factor combinations and industry rotation strategies, with notable differentiation across different stock pools[3][5][7] - **Model Name**: Barra **Model Construction Idea**: Barra focuses on style factors to explain stock returns and risks[14][15][16] **Model Construction Process**: Barra includes multiple style factors such as Beta, Size, Momentum, Volatility, Non-linear Size, Valuation, Liquidity, Profitability, Growth, and Leverage. Each factor is calculated using specific formulas: - **Beta**: Historical beta - **Size**: Natural logarithm of total market capitalization - **Momentum**: Mean of historical excess return series - **Volatility**: $0.74 \times \text{historical excess return volatility} + 0.16 \times \text{cumulative excess return deviation} + 0.1 \times \text{historical residual return volatility}$ - **Non-linear Size**: Cubic transformation of market capitalization - **Valuation**: Reciprocal of price-to-book ratio - **Liquidity**: $0.35 \times \text{monthly turnover rate} + 0.35 \times \text{quarterly turnover rate} + 0.3 \times \text{annual turnover rate}$ - **Profitability**: Weighted combination of analyst forecast earnings-price ratio, reciprocal of cash flow ratio, reciprocal of trailing twelve-month P/E 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] **Model Evaluation**: Barra style factors provide a comprehensive framework for analyzing stock returns, with strong differentiation in multi-factor strategies[14][15][16] Model Backtesting Results - **GRU Model**: - **open1d**: Weekly excess return 0.61%, monthly 1.56%, yearly 7.78% - **close1d**: Weekly excess return 0.02%, monthly 1.45%, yearly 7.28% - **barra1d**: Weekly excess return -0.24%, monthly -0.07%, yearly 3.61% - **barra5d**: Weekly excess return 0.06%, monthly 1.35%, yearly 8.63% - **Multi-factor combination**: Weekly excess return 0.61%, monthly 0.82%, yearly 3.22%[31][32][33] Quantitative Factors and Construction Methods - **Factor Name**: Beta **Factor Construction Idea**: Measures historical sensitivity to market movements[15] **Factor Construction Process**: Calculated as historical beta using regression analysis of stock returns against market returns[15] - **Factor Name**: Size **Factor Construction Idea**: Captures the impact of market capitalization on stock returns[15] **Factor Construction Process**: Natural logarithm of total market capitalization[15] - **Factor Name**: Momentum **Factor Construction Idea**: Reflects the persistence of stock price trends[15] **Factor Construction Process**: Mean of historical excess return series[15] - **Factor Name**: Volatility **Factor Construction Idea**: Measures risk associated with stock price fluctuations[15] **Factor Construction Process**: $0.74 \times \text{historical excess return volatility} + 0.16 \times \text{cumulative excess return deviation} + 0.1 \times \text{historical residual return volatility}$[15] - **Factor Name**: Non-linear Size **Factor Construction Idea**: Captures non-linear effects of market capitalization on returns[15] **Factor Construction Process**: Cubic transformation of market capitalization[15] - **Factor Name**: Valuation **Factor Construction Idea**: Reflects the relative attractiveness of stock prices[15] **Factor Construction Process**: Reciprocal of price-to-book ratio[15] - **Factor Name**: Liquidity **Factor Construction Idea**: Measures ease of trading stocks[15] **Factor Construction Process**: $0.35 \times \text{monthly turnover rate} + 0.35 \times \text{quarterly turnover rate} + 0.3 \times \text{annual turnover rate}$[15] - **Factor Name**: Profitability **Factor Construction Idea**: Captures earnings quality and growth potential[15] **Factor Construction Process**: Weighted combination of analyst forecast earnings-price ratio, reciprocal of cash flow ratio, reciprocal of trailing twelve-month P/E ratio, and forecasted growth rates[15] - **Factor Name**: Growth **Factor Construction Idea**: Reflects revenue and earnings growth trends[15] **Factor Construction Process**: Weighted combination of earnings growth rate and revenue growth rate[15] - **Factor Name**: Leverage **Factor Construction Idea**: Measures financial risk associated with debt levels[15] **Factor Construction Process**: Weighted combination of market leverage, book leverage, and debt-to-asset ratio[15] Factor Backtesting Results - **Beta**: Weekly excess return -0.24%, monthly -0.07%, yearly 3.61%[31][32][33] - **Size**: Weekly excess return 0.02%, monthly 1.45%, yearly 7.28%[31][32][33] - **Momentum**: Weekly excess return 0.61%, monthly 1.56%, yearly 7.78%[31][32][33] - **Volatility**: Weekly excess return 0.06%, monthly 1.35%, yearly 8.63%[31][32][33] - **Non-linear Size**: Weekly excess return 0.61%, monthly 0.82%, yearly 3.22%[31][32][33] - **Valuation**: Weekly excess return 0.61%, monthly 0.82%, yearly 3.22%[31][32][33] - **Liquidity**: Weekly excess return 0.61%, monthly 0.82%, yearly 3.22%[31][32][33] - **Profitability**: Weekly excess return 0.61%, monthly 0.82%, yearly 3.22%[31][32][33] - **Growth**: Weekly excess return 0.61%, monthly 0.82%, yearly 3.22%[31][32][33] - **Leverage**: Weekly excess return 0.61%, monthly 0.82%, yearly 3.22%[31][32][33]

Quantitative Models and Construction 1. Model Name: GRU Model - **Model Construction Idea**: The GRU model integrates fundamental and technical features to predict stock performance, leveraging historical data and recurrent neural network structures for time-series analysis [3][4][5]. - **Model Construction Process**: - Input features include fundamental indicators (e.g., financial ratios) and technical indicators (e.g., momentum, volatility) [3][4]. - The GRU (Gated Recurrent Unit) architecture processes sequential data to capture temporal dependencies [3]. - The model is trained on historical data, with optimization targeting the minimization of prediction errors [3]. - **Model Evaluation**: The GRU model shows mixed performance across different stock pools, with notable underperformance in certain scenarios [3][5][6]. 2. Model Name: Barra1d - **Model Construction Idea**: The Barra1d model is a factor-based model that emphasizes short-term price movements and volatility [3][4][5]. - **Model Construction Process**: - Factors include short-term momentum and volatility metrics [3][4]. - The model applies a linear regression framework to estimate factor exposures and returns [3]. - Portfolio construction involves long positions in stocks with high factor scores and short positions in stocks with low scores [3][4]. - **Model Evaluation**: Barra1d demonstrates strong performance in multiple stock pools, with consistent positive returns in backtests [4][5][6]. 3. Model Name: Barra5d - **Model Construction Idea**: Barra5d extends the Barra1d model by incorporating a longer time horizon for factor evaluation [3][4][5]. - **Model Construction Process**: - Factors include medium-term momentum and volatility metrics [3][4]. - The model uses a similar regression-based approach as Barra1d but adjusts for longer-term trends [3]. - Portfolio construction follows the same long-short strategy as Barra1d [3][4]. - **Model Evaluation**: Barra5d shows strong year-to-date performance, outperforming benchmarks in multiple scenarios [5][6][7]. --- Model Backtest Results GRU Model - **Close1d**: Weekly return -1.59%, YTD return 8.61% [31] - **Barra1d**: Weekly return 0.80%, YTD return 22.50% [31] - **Barra5d**: Weekly return 0.63%, YTD return 28.18% [31] Multi-Factor Portfolio - Weekly excess return: -0.19% - YTD excess return: 2.73% [34] --- Quantitative Factors and Construction 1. Factor Name: Beta - **Factor Construction Idea**: Measures historical sensitivity of stock returns to market movements [15]. - **Factor Construction Process**: - Calculated as the slope of the regression of stock returns against market returns over a specified period [15]. - **Factor Evaluation**: Beta factor showed strong long-side performance in recent weeks [16]. 2. Factor Name: Momentum - **Factor Construction Idea**: Captures the persistence of stock price trends [15]. - **Factor Construction Process**: - Calculated as the mean of historical excess returns over a defined period [15]. - **Factor Evaluation**: Long-term momentum factors demonstrated positive returns, while short-term momentum factors underperformed [18][20]. 3. Factor Name: Volatility - **Factor Construction Idea**: Measures the variability of stock returns [15]. - **Factor Construction Process**: - Weighted combination of historical return volatility, cumulative deviation, and residual volatility [15]. - **Factor Evaluation**: Volatility factors showed strong positive returns, particularly in long-term horizons [18][20]. 4. Factor Name: Growth - **Factor Construction Idea**: Reflects the growth potential of companies based on financial metrics [15]. - **Factor Construction Process**: - Weighted combination of earnings growth rate and revenue growth rate [15]. - **Factor Evaluation**: Growth factors exhibited strong positive returns across multiple stock pools [18][20][23]. --- Factor Backtest Results Beta Factor - Weekly return: Positive [16] Momentum Factor - Long-term momentum: Weekly return 2.87% [22] - Short-term momentum: Weekly return -3.48% [22] Volatility Factor - Long-term volatility: Weekly return 4.01% [22] - Short-term volatility: Weekly return 2.87% [22] Growth Factor - Weekly return: Positive [18][20][23]

Quantitative Models and Factor Construction Factor Selection and Data Processing Pipeline - The MSCI Barra CNE5 model includes 10 primary factors and 21 secondary factors, covering classic academic factors such as beta, size, and book-to-price ratio, as well as fundamental and technical factors like value, growth, momentum, and residual volatility[22][23][24] - Secondary factors are standardized and weighted to synthesize primary factors, with weights optimized for explanatory power. However, later versions of MSCI Barra shifted to equal weighting for simplicity[23] - Data processing pipeline includes six steps: defining the base universe, outlier handling, missing value imputation, standardization, primary factor synthesis, and secondary outlier/standardization adjustments[31][32][35] Pure Factor Return Estimation - Pure factor returns are estimated using constrained weighted least squares (WLS). Constraints are introduced to address multicollinearity caused by the inclusion of intercepts (country factors)[44][45][49] - WLS weights are inversely proportional to the square root of market capitalization, ensuring smaller residual variance for larger stocks[45] - The solution for pure factor returns is derived using matrix transformations and Cholesky decomposition, ensuring variance homogeneity[46][57][59] Evaluation of Risk Factors and Factor Systems - MSCI Barra's six-dimensional evaluation criteria include statistical significance, stability, intuition, completeness, simplicity, and low multicollinearity[75][76][77] - Quantitative metrics such as average absolute t-values, variance inflation factors (VIF), and pure factor performance are used to assess factor quality. Factors like beta, liquidity, and size exhibit strong statistical significance but may overlap in information[83][84][85] Practical Applications of Risk Models - Risk models are applied for performance attribution in external products (e.g., public equity funds) and internal portfolios (e.g., brokerage "gold stock" portfolios). Attribution results include style/sector exposures and return/risk contributions[148][151][181] - For public equity funds, factor and idiosyncratic returns are decomposed to classify funds into "style advantage" or "stock-picking advantage" categories[152][153][155] - For brokerage gold stock portfolios, attribution reveals the superior performance of newly added stocks due to idiosyncratic returns, while recent underperformance is linked to systematic exposure to small-cap factors[157][169][170] --- Factor Backtesting Results Daily Frequency Results - **Beta**: Annual return 8.20%, annual volatility 4.87%, IR 1.69[86][111] - **Size**: Annual return -6.82%, annual volatility 4.57%, IR -1.49[86][105] - **Liquidity**: Annual return -9.46%, annual volatility 3.10%, IR -3.05[86][123] - **Value**: Annual return 4.32%, annual volatility 2.40%, IR 1.80[86][134] Monthly Frequency Results - **Beta**: Annual return 2.64%, annual volatility 3.95%, IR 0.15[95][111] - **Size**: Annual return -7.02%, annual volatility 5.99%, IR -0.26[95][105] - **Liquidity**: Annual return -5.74%, annual volatility 2.77%, IR -0.45[95][123] - **Value**: Annual return 2.94%, annual volatility 2.87%, IR 0.22[95][134] Gold Stock Portfolio Attribution - **All Gold Stocks**: Total return 61.86%, factor return -54.02%, idiosyncratic return 83.46%[171] - **Newly Added Gold Stocks**: Total return 83.50%, factor return -59.75%, idiosyncratic return 108.20%[174] - **Repeated Gold Stocks**: Total return 6.39%, factor return -44.66%, idiosyncratic return 19.60%[162] Factor Contribution Analysis - **Beta**: Positive contribution across all years, cumulative return 35.75% for all gold stocks, 44.47% for newly added gold stocks[175][176] - **Liquidity**: Negative contribution, cumulative return -48.67% for all gold stocks, -57.24% for newly added gold stocks[175][176] - **Size**: Mixed contribution, cumulative return 72.78% for all gold stocks, 97.27% for newly added gold stocks[175][176]

Group 1 - The core view of the report indicates that risk assets are expected to maintain relative advantages, with the profitability factor likely to recover [2][4][10] - The asset allocation model is an improved version of the Black-Litterman (BL) model, which combines market consensus with active views to optimize asset allocation and enhance the Sharpe ratio [3][5] - The model predicts that in the third quarter of 2025, the allocation ratio for domestic stocks will continue to increase while the bond allocation ratio will remain relatively high [10][11] Group 2 - In the A-share market, the profitability factor is expected to recover, and the advantage of small-cap stocks is likely to continue [2][17] - As of June 30, 2025, the market style performance for the second quarter showed strong results for small-cap and low-valuation factors, with weak profitability and weak reversal [13][16] - The report recommends focusing on indices such as the ChiNext Index, CSI A500, and CSI 2000, which exhibit high profitability and small-cap attributes [20][21]

Quantitative Models and Construction - **Model Name**: barra1d **Model Construction Idea**: Focuses on short-term factor performance using daily data **Model Construction Process**: Utilizes historical data to calculate factor exposures and applies industry-neutral adjustments. Stocks are ranked based on factor scores, with the top 10% selected for long positions and the bottom 10% for short positions. Adjustments include equal weighting and monthly rebalancing[19][21][30] **Model Evaluation**: Demonstrates strong performance in short-term factor analysis[19][21][30] - **Model Name**: barra5d **Model Construction Idea**: Focuses on medium-term factor performance using five-day data **Model Construction Process**: Similar to barra1d, but uses a five-day rolling window for factor calculations. Stocks are ranked and selected based on factor scores, with monthly rebalancing and equal weighting applied[19][21][30] **Model Evaluation**: Exhibits robust medium-term factor performance, outperforming other models in cumulative returns[19][21][30] - **Model Name**: open1d **Model Construction Idea**: Focuses on factor performance using daily open prices **Model Construction Process**: Factors are calculated using daily open price data, with industry-neutral adjustments applied. Stocks are ranked based on factor scores, and the top 10% are selected for long positions, while the bottom 10% are shorted. Monthly rebalancing is implemented[19][21][30] **Model Evaluation**: Performs well in certain market conditions but shows higher volatility compared to other models[19][21][30] - **Model Name**: close1d **Model Construction Idea**: Focuses on factor performance using daily close prices **Model Construction Process**: Factors are calculated using daily close price data, with industry-neutral adjustments applied. Stocks are ranked based on factor scores, and the top 10% are selected for long positions, while the bottom 10% are shorted. Monthly rebalancing is implemented[19][21][30] **Model Evaluation**: Demonstrates weaker performance compared to other models, with significant drawdowns observed[19][21][30] Model Backtesting Results - **barra1d**: Weekly excess return 0.17%, monthly excess return 0.32%, six-month excess return 4.09%, year-to-date excess return 3.93%[32] - **barra5d**: Weekly excess return 0.13%, monthly excess return 0.39%, six-month excess return 7.59%, year-to-date excess return 7.56%[32] - **open1d**: Weekly excess return -0.35%, monthly excess return -0.71%, six-month excess return 5.85%, year-to-date excess return 6.30%[32] - **close1d**: Weekly excess return 0.55%, monthly excess return 0.40%, six-month excess return 6.40%, year-to-date excess return 6.31%[32] - **Multi-factor model**: Weekly excess return -0.38%, monthly excess return -0.04%, six-month excess return 3.56%, year-to-date excess return 2.82%[32] Quantitative Factors and Construction - **Factor Name**: Beta **Factor Construction Idea**: Measures historical beta to assess market sensitivity **Factor Construction Process**: Calculated using historical beta values derived from regression analysis of stock returns against market returns[15][16] **Factor Evaluation**: Demonstrates strong performance in high-volatility environments[15][16] - **Factor Name**: Momentum **Factor Construction Idea**: Captures historical excess return trends **Factor Construction Process**: Combines weighted averages of historical excess return volatility, cumulative excess return deviation, and residual return volatility using the formula: $ Momentum = 0.74 * Historical Excess Return Volatility + 0.16 * Cumulative Excess Return Deviation + 0.1 * Residual Return Volatility $[15][16] **Factor Evaluation**: Performs well in trending markets but struggles in reversal scenarios[15][16] - **Factor Name**: Volatility **Factor Construction Idea**: Measures stock price fluctuation intensity **Factor Construction Process**: Combines weighted averages of monthly, quarterly, and annual turnover rates using the formula: $ Volatility = 0.35 * Monthly Turnover Rate + 0.35 * Quarterly Turnover Rate + 0.3 * Annual Turnover Rate $[15][16] **Factor Evaluation**: Strong performance in high-volatility stocks[15][16] - **Factor Name**: Valuation **Factor Construction Idea**: Assesses stock valuation using price-to-book ratio **Factor Construction Process**: Calculated as the inverse of the price-to-book ratio[15][16] **Factor Evaluation**: Performs well in identifying undervalued stocks[15][16] Factor Backtesting Results - **Beta**: Weekly excess return 0.17%, monthly excess return 0.32%, six-month excess return 4.09%, year-to-date excess return 3.93%[32] - **Momentum**: Weekly excess return -0.38%, monthly excess return -0.04%, six-month excess return 3.56%, year-to-date excess return 2.82%[32] - **Volatility**: Weekly excess return 0.55%, monthly excess return 0.40%, six-month excess return 6.40%, year-to-date excess return 6.31%[32] - **Valuation**: Weekly excess return 0.13%, monthly excess return 0.39%, six-month excess return 7.59%, year-to-date excess return 7.56%[32]