Core Viewpoint - The article provides an analysis of market performance, highlighting the positive and negative returns of various factors across different stock pools and industries, indicating a mixed market sentiment and specific investment opportunities [2][3][4]. Group 1: Market Factor Performance - In the overall market stock pool, the Beta factor and valuation factor achieved positive returns of 0.48% and 0.26% respectively, while the market capitalization factor and profit factor recorded negative returns of -0.24% and -0.42%, suggesting a small-cap style market [2]. - The momentum factor yielded a negative return of -0.44%, indicating a reversal style in the market [2]. Group 2: Single Factor Performance - In the CSI 300 stock pool, the best-performing factors included quarterly net profit year-on-year growth rate (1.83%), quarterly operating profit year-on-year growth rate (1.75%), and net profit margin TTM (1.52%) [3]. - In the CSI 500 stock pool, the top factors were price-to-book ratio (2.57%), downside volatility ratio (2.07%), and inverse of price-to-sales ratio TTM (1.33%) [3]. - In the liquidity 1500 stock pool, the leading factors were downside volatility ratio (0.74%), net profit gap (0.49%), and quarterly ROE year-on-year (0.46%) [3]. Group 3: Industry Factor Performance - The fundamental factors showed varied performance across industries, with net asset growth rate, net profit growth rate, earnings per share, and operating profit TTM factors yielding consistent positive returns in the transportation industry [4]. - Among valuation factors, the BP factor performed well, showing significant positive returns in the real estate industry [4]. - Residual volatility and liquidity factors demonstrated notable positive returns in the non-ferrous metals industry [4]. Group 4: Investment Strategy Performance - The PB-ROE-50 combination achieved positive excess returns in the overall market stock pool, while it recorded excess returns of -0.56% in the CSI 500 stock pool and -0.38% in the CSI 800 stock pool [5]. - Public and private fund research selection strategies both gained positive excess returns, with public strategies outperforming the CSI 800 by 0.80% and private strategies by 1.21% [6]. - The block trading combination also achieved positive excess returns relative to the CSI All Index, with an excess return of 1.22% [7]. - The targeted issuance combination gained positive excess returns relative to the CSI All Index, with an excess return of 0.05% [8].

Quantitative Models and Construction Methods 1. Model Name: PB-ROE-50 - **Model Construction Idea**: The PB-ROE-50 model selects stocks based on a combination of Price-to-Book (PB) ratio and Return on Equity (ROE), aiming to capture value and profitability factors[23] - **Model Construction Process**: - Stocks are ranked based on their PB and ROE metrics - A portfolio is constructed by selecting the top 50 stocks with the best combined PB and ROE scores - The portfolio is rebalanced periodically to maintain the factor exposure[23] - **Model Evaluation**: The model demonstrates the ability to generate excess returns in certain market conditions, particularly in capturing value and profitability factors[23] 2. Model Name: Block Trade Portfolio - **Model Construction Idea**: This model leverages the information embedded in block trades, focusing on stocks with high block trade transaction amounts and low volatility in transaction amounts[29] - **Model Construction Process**: - Identify stocks with high "block trade transaction amount ratio" and low "6-day transaction amount volatility" - Construct a portfolio based on these criteria and rebalance monthly[29] - **Model Evaluation**: The model effectively captures the excess return potential of block trade-related stocks, particularly those with high transaction amounts and low volatility[29] 3. Model Name: Private Placement Portfolio - **Model Construction Idea**: This model focuses on stocks involved in private placements, aiming to capture the event-driven effects of private placements on stock performance[35] - **Model Construction Process**: - Use the shareholder meeting announcement date as the event trigger - Incorporate market capitalization, rebalancing frequency, and position control into the portfolio construction process - Construct a portfolio based on these parameters[35] - **Model Evaluation**: The model captures the investment opportunities arising from private placement events, though its effectiveness may vary depending on market conditions[35] --- Model Backtesting Results 1. PB-ROE-50 Model - **Excess Return (This Week)**: - CSI 500: -0.56% - CSI 800: -0.38% - All Market: 0.92%[24] - **Year-to-Date Excess Return**: - CSI 500: 2.99% - CSI 800: 6.41% - All Market: 9.28%[24] - **Absolute Return (This Week)**: - CSI 500: 1.39% - CSI 800: 0.73% - All Market: 2.47%[24] - **Year-to-Date Absolute Return**: - CSI 500: 8.41% - CSI 800: 9.44% - All Market: 16.07%[24] 2. Block Trade Portfolio - **Excess Return (This Week)**: 1.22% - **Year-to-Date Excess Return**: 25.89% - **Absolute Return (This Week)**: 2.78% - **Year-to-Date Absolute Return**: 33.71%[30] 3. Private Placement Portfolio - **Excess Return (This Week)**: 0.05% - **Year-to-Date Excess Return**: 8.72% - **Absolute Return (This Week)**: 1.59% - **Year-to-Date Absolute Return**: 15.48%[36] --- Quantitative Factors and Construction Methods 1. Factor Name: Beta Factor - **Factor Construction Idea**: Measures the sensitivity of a stock's returns to market returns, capturing systematic risk[18] - **Factor Construction Process**: - Calculate the covariance between the stock's returns and market returns - Divide by the variance of market returns to derive the beta coefficient - $ \beta = \frac{\text{Cov}(R_i, R_m)}{\text{Var}(R_m)} $ - Where $R_i$ is the stock return, and $R_m$ is the market return[18] - **Factor Evaluation**: The factor captures systematic risk effectively and is widely used in portfolio construction and risk management[18] 2. Factor Name: Momentum Factor - **Factor Construction Idea**: Measures the tendency of stocks with high past returns to continue performing well in the future[18] - **Factor Construction Process**: - Calculate the cumulative return of a stock over a specific look-back period (e.g., 6 months or 12 months) - Rank stocks based on their cumulative returns and construct a portfolio of top-ranked stocks[18] - **Factor Evaluation**: The factor is effective in capturing trends in stock performance but may underperform in reversal markets[18] 3. Factor Name: Valuation Factor (e.g., PB, PE, PS) - **Factor Construction Idea**: Measures the relative valuation of stocks based on financial metrics like Price-to-Book (PB), Price-to-Earnings (PE), and Price-to-Sales (PS)[18] - **Factor Construction Process**: - Calculate the PB, PE, or PS ratio for each stock - Rank stocks based on these ratios and construct a portfolio of low-ratio stocks (value stocks)[18] - **Factor Evaluation**: Valuation factors are effective in identifying undervalued stocks but may underperform during growth-driven market phases[18] --- Factor Backtesting Results 1. Beta Factor - **Weekly Return**: 0.48%[18] 2. Momentum Factor - **Weekly Return**: -0.44%[18] 3. Valuation Factor - **Weekly Return**: - PB: 2.57% (CSI 500)[14] - PE: 0.37% (CSI 300)[13] - PS: 1.26% (CSI 300)[13]

Core Viewpoint - The report highlights the performance of various investment factors and strategies over the week, indicating a mixed market environment with specific factors yielding positive and negative returns [3][4][5]. Factor Performance - BP factor and profit factor achieved positive returns of 0.30% and 0.27% respectively, while non-linear market capitalization factor and size factor showed significant negative returns of -0.31% and -0.29%, indicating a clear small-cap market style [3]. - In the CSI 300 stock pool, the best-performing factors included TTM P/E ratio (0.70%), TTM P/S ratio (0.59%), and 5-minute return skewness (0.57%), while the worst performers were 6-day trading volume moving average (-1.24%), 5-day average turnover rate (-1.44%), and TTM gross profit margin (-1.62%) [4]. - In the CSI 500 stock pool, the top factors were quarterly ROE (1.70%), TTM gross profit margin (1.54%), and 5-day trading volume standard deviation (1.36%), with poor performers including early trading return factor (-0.39%), 5-minute return skewness (-0.44%), and log market capitalization factor (-0.73%) [4]. - In the liquidity 1500 stock pool, the best factors were 5-day reversal (1.62%), quarterly ROE (1.53%), and P/E factor (1.41%), while the worst were 6-day trading volume moving average (-0.61%), early trading return factor (-0.70%), and 5-day index moving average of trading volume (-0.72%) [4]. Industry Factor Performance - The net asset growth rate factor showed significant positive returns in the comprehensive industry, while the net profit growth rate factor performed well across the same sector [5]. - The 5-day momentum factor exhibited strong momentum effects in the comprehensive, steel, and public utility industries, while reversal effects were notable in non-bank financials, non-ferrous metals, and telecommunications [5]. - Valuation factors like BP factor performed well in the comprehensive, steel, and banking industries, while EP factor excelled in the comprehensive, media, and non-bank financial sectors [5]. Strategy Performance - The PB-ROE-50 combination achieved excess returns across various stock pools, with excess returns of 1.17% in the CSI 500 pool, 1.21% in the CSI 800 pool, and 1.36% in the overall market stock pool [6]. - Public fund research selection strategy and private fund research tracking strategy both gained positive excess returns, with public fund strategy achieving 0.02% excess return relative to CSI 800 and private fund strategy achieving 0.25% [7]. - The block trading combination experienced a relative excess return drawdown of -0.24% compared to the CSI All Index [8]. - The directed issuance combination also faced a relative excess return drawdown of -0.69% compared to the CSI All Index [9].

Quantitative Models and Construction Methods - **Model Name**: Neutral Constraint Maximum Factor Exposure Portfolio **Model Construction Idea**: The model aims to maximize the exposure of the target factor in the portfolio while maintaining neutrality in industry and style exposures relative to the benchmark index[60][61] **Model Construction Process**: 1. The objective function is to maximize the portfolio's exposure to the target factor $ \text{Max} \quad w^{\prime} X_{\text{target}} $ 2. Constraints include: - Industry neutrality: $ (w - w_b)^{\prime} X_{\text{inad}} = 0 $ - Style neutrality: $ (w - w_b)^{\prime} X_{\text{Beta}} = 0 $ - Weight deviation limit: $ |w - w_b| \leq 1\% $ - No short selling: $ w \geq 0 $ - Full investment: $ w^{\prime} B = 1 $ and $ w^{\prime} 1 = 1 $ 3. Factors are neutralized to remove correlations with industry and style factors (e.g., size, valuation, growth) 4. Factor directions are adjusted to be positive before optimization[60][62][63] **Model Evaluation**: The model effectively balances factor exposure maximization with risk control, ensuring alignment with the benchmark index[63] --- Quantitative Factors and Construction Methods - **Factor Name**: Valuation Factor (BP) **Factor Construction Idea**: Captures the valuation level of stocks based on book-to-price ratio[14][15] **Factor Construction Process**: - Formula: $ \text{BP} = \frac{\text{Shareholders' Equity}}{\text{Market Capitalization}} $ **Factor Evaluation**: Demonstrates strong performance in capturing undervalued stocks, particularly in defensive market environments[14][15] - **Factor Name**: Growth Factor **Factor Construction Idea**: Measures the growth potential of stocks based on sales and earnings growth rates[14][15] **Factor Construction Process**: - Formula: $ \text{Growth Factor} = \frac{\text{SGRO} + \text{EGRO}}{2} $ - $ \text{SGRO} $: Regression slope of past 5 years' annual sales divided by average sales - $ \text{EGRO} $: Regression slope of past 5 years' annual earnings divided by average earnings[14][15] **Factor Evaluation**: Useful in identifying high-growth stocks, though performance may vary across market cycles[14][15] - **Factor Name**: Beta Factor **Factor Construction Idea**: Reflects the sensitivity of a stock's returns to market movements[14][15] **Factor Construction Process**: - Formula: $ \text{Beta} = \text{Regression Coefficient of Stock Returns on Market Returns} $ - Regression uses 252 trading days with a half-life of 63 days[14][15] **Factor Evaluation**: Effective in capturing market risk preferences, with low-beta stocks outperforming in risk-averse environments[14][15] - **Factor Name**: Momentum Factor (RSTR) **Factor Construction Idea**: Identifies stocks with strong relative strength over a specific period[14][15] **Factor Construction Process**: - Formula: $ \text{RSTR} = \text{Cumulative Returns over 504 Days (Excluding Last 21 Days)} $ - Returns are weighted using a half-life of 126 days[14][15] **Factor Evaluation**: Performs well in trending markets but may underperform during reversals[14][15] --- Factor Backtesting Results - **Valuation Factor (BP)**: - Recent 1-week long-short return: 1.90% - Recent 1-month long-short return: -0.58%[17] - **Growth Factor**: - Recent 1-week long-short return: -0.79% - Recent 1-month long-short return: -0.99%[17] - **Beta Factor**: - Recent 1-week long-short return: -2.89% - Recent 1-month long-short return: 5.63%[17] - **Momentum Factor (RSTR)**: - Recent 1-week long-short return: -1.03% - Recent 1-month long-short return: -2.31%[17] --- Index Enhancement Portfolio Backtesting Results - **CSI 300 Enhanced Portfolio**: - 1-week excess return: 0.63% - 1-month excess return: 2.22% - 1-year excess return: 2.43%[57][58] - **CSI 500 Enhanced Portfolio**: - 1-week excess return: 0.10% - 1-month excess return: -1.25% - 1-year excess return: -2.90%[57][58] - **CSI 800 Enhanced Portfolio**: - 1-week excess return: 0.70% - 1-month excess return: 1.47% - 1-year excess return: 1.03%[57][58] - **CSI 1000 Enhanced Portfolio**: - 1-week excess return: 0.16% - 1-month excess return: 0.21% - 1-year excess return: 12.99%[57][58] - **CSI 300 ESG Enhanced Portfolio**: - 1-week excess return: 0.36% - 1-month excess return: 2.64% - 1-year excess return: 7.88%[57][58]

- The Beta factor showed a significant positive return of 6.95% this week, indicating a strong market preference for high Beta stocks [10] - The Liquidity factor also performed well with a return of 5.53%, reflecting increased demand for highly liquid assets [10] - The Volatility factor improved significantly with a return of 4.19%, showing heightened market interest in high-volatility assets [10] - The Trend factor experienced a notable decline, with a return of -1.76%, indicating a reduced market preference for trend-following strategies [11] - The Size factor showed a significant drop with a return of -3.30%, indicating a decreased market focus on small-cap stocks [11] - The Value factor also declined sharply, with a return of -3.55%, reflecting a reduced market preference for value investment strategies [11] - The one-year momentum factor performed well across various indices, including the CSI 500 and CSI 1000, indicating strong performance in the past year [7][24][30] - The DELTAROE factor showed strong performance in indices like the CSI 800 and CSI 2000, indicating robust profitability growth [27][33] - The three-month reversal factor also performed well in multiple indices, reflecting a strong short-term reversal trend [7][24][27] - The UMR factors, including one-month, three-month, and six-month UMR, generally performed poorly across various indices, indicating weak momentum [7][24][27][30] - The public fund index enhancement products for the CSI 300, CSI 500, and CSI 1000 showed varying levels of excess returns, with the CSI 300 products generally outperforming the others [7][46][48][50] - The MFE (Maximized Factor Exposure) portfolio construction method was used to evaluate the effectiveness of individual factors under various constraints, ensuring controlled industry and style exposures [51][52][54][55]

Quantitative Models and Construction Methods - **Model Name**: Neutral Constraint Maximum Factor Exposure Portfolio **Construction Idea**: Maximize the exposure of the target factor in the portfolio while maintaining neutrality in industry and style exposures relative to the benchmark index[60][61][63] **Construction Process**: 1. **Objective Function**: Maximize portfolio exposure to the target factor $ \text{Max}\quad w^{\prime}X_{\text{target}} $ 2. **Constraints**: - Industry neutrality: $ (w - w_b)^{\prime}X_{\text{inad}} = 0 $ - Style neutrality: $ (w - w_b)^{\prime}X_{\text{Beta}} = 0 $ - Weight deviation limit: $ |w - w_b| \leq 1\% $ - No short selling: $ w \geq 0 $ - Full allocation: $ w^{\prime}1 = 1 $ - Constituents from benchmark index: $ w^{\prime}B = 1 $ **Evaluation**: The model ensures that the portfolio remains neutral to industry and style biases while maximizing factor exposure[60][61][63] Factor Construction and Definitions - **Factor Name**: Beta Factor **Construction Idea**: Capture the sensitivity of individual stock returns to market returns[14][15] **Construction Process**: - Calculate the regression coefficient of individual stock daily returns against the market index (CSI All Share Index) over the past 252 trading days using a half-life weighting of 63 days **Formula**: $ \text{Beta} = \text{Regression Coefficient} $ **Evaluation**: Reflects market risk sensitivity, useful for identifying high-risk or low-risk stocks[14][15] - **Factor Name**: Book-to-Price (BP) **Construction Idea**: Measure valuation by comparing book equity to market capitalization[14][15] **Construction Process**: - $ \text{BP} = \frac{\text{Shareholders' Equity}}{\text{Market Capitalization}} $ **Evaluation**: Indicates undervaluation or overvaluation of stocks, commonly used in value investing[14][15] - **Factor Name**: Sales Growth (SGRO) **Construction Idea**: Assess growth potential by analyzing historical revenue trends[14][15] **Construction Process**: - Perform regression on annual revenue data from the past five fiscal years - Divide the regression slope by the average revenue to calculate growth rate **Formula**: $ \text{SGRO} = \frac{\text{Regression Slope}}{\text{Average Revenue}} $ **Evaluation**: Useful for identifying companies with strong growth trajectories[14][15] Factor Backtesting Results - **Beta Factor**: Weekly long-short return of 7.50%, monthly return of 8.74%[16] - **Book-to-Price (BP)**: Weekly return of -0.27%, monthly return of 0.39%[21][26][30] - **Sales Growth (SGRO)**: Not explicitly tested in the report[14][15] Portfolio Backtesting Results - **Neutral Constraint Maximum Factor Exposure Portfolio**: - **CSI 300 Enhanced Portfolio**: Weekly excess return of 0.03%, monthly return of 1.91%, annual return of 1.34%[57][58] - **CSI 500 Enhanced Portfolio**: Weekly excess return of -1.29%, monthly return of -1.24%, annual return of -2.54%[57][58] - **CSI 800 Enhanced Portfolio**: Weekly excess return of -0.32%, monthly return of 1.68%, annual return of 1.19%[57][58] - **CSI 1000 Enhanced Portfolio**: Weekly excess return of -0.95%, monthly return of 1.33%, annual return of 13.01%[57][58] - **CSI 300 ESG Enhanced Portfolio**: Weekly excess return of 0.51%, monthly return of 2.44%, annual return of 7.72%[57][58] Factor Performance in Different Stock Pools - **CSI 300 Stock Pool**: - Weekly top-performing factors: Log Market Cap (0.83%), Single Quarter Operating Profit Growth (0.72%), 20-Day Specificity (0.71%)[21][23] - Monthly top-performing factors: Single Quarter EP (3.19%), EP_TTM (2.93%), Single Quarter ROE (2.63%)[24] - **CSI 500 Stock Pool**: - Weekly top-performing factors: 20-Day Specificity (1.39%), 60-Day Volume Ratio (1.13%), 60-Day Reversal (1.00%)[26][28] - Monthly top-performing factors: Single Quarter Revenue Growth (3.31%), Single Quarter Operating Profit Growth (2.73%), Single Quarter ROE Growth (2.72%)[28] - **CSI 800 Stock Pool**: - Weekly top-performing factors: Log Market Cap (1.59%), Single Quarter ROE Growth (1.20%), Single Quarter Operating Profit Growth (1.06%)[30][32] - Monthly top-performing factors: Single Quarter EP (4.36%), Single Quarter ROE Growth (3.90%), Single Quarter ROE (3.90%)[33] - **CSI 1000 Stock Pool**: - Weekly top-performing factors: 60-Day Reversal (1.40%), Single Quarter SP (1.30%), SP_TTM (1.29%)[35][37] - Monthly top-performing factors: Log Market Cap (3.66%), 60-Day Reversal (3.43%), Single Quarter Net Profit Growth (3.24%)[38] - **CSI 300 ESG Stock Pool**: - Weekly top-performing factors: Log Market Cap (1.05%), 20-Day Volume Ratio (0.63%), 20-Day Specificity (0.60%)[40][41] - Monthly top-performing factors: Log Market Cap (4.20%), Single Quarter ROE (2.55%), EP_TTM (2.49%)[42] - **All-Market Stock Pool**: - Weekly top-performing factors: Log Market Cap (24.81% Rank IC), 20-Day Specificity (21.07% Rank IC), 60-Day Reversal (19.50% Rank IC)[44][45] - Monthly top-performing factors: 20-Day Specificity (11.25% Rank IC), 20-Day Three-Factor Model Residual Volatility (10.96% Rank IC), 60-Day Specificity (10.73% Rank IC)[45]

Core Viewpoint - The article emphasizes the importance of effectively controlling risks and reducing volatility in asset management, advocating for a "macro factor" risk parity strategy that adapts to changing macroeconomic environments, contrasting it with traditional asset risk parity models [1][13][15]. Group 1: Macro Factor Risk Parity Framework - The construction of a macro factor risk parity framework involves four steps: selecting factors, calculating risk exposure, determining target risk exposure, and matching target risk exposure to asset weights [2][16][17]. - The mainstream methods for constructing macro factors include using low-frequency economic data, principal component analysis (PCA), and regression methods to fit higher-frequency macro factors [3][18][19]. Group 2: Factor Construction and High-Frequency Transformation - The article outlines a refined approach to factor construction, summarizing it as "defining dimensions, screening assets, and high-frequency transformation," which combines the advantages of various methods [3][18][19]. - The transformation of low-frequency macro factors into high-frequency factors is achieved through factor mimicking, which involves regression analysis to identify strong correlations with asset prices [5][29][31]. Group 3: Risk Exposure and Asset Sensitivity - A risk exposure matrix is created to show the sensitivity of assets to different macro variables, using robust OLS regression to capture dynamic features [6][33][34]. - The analysis reveals that large-cap stocks are more sensitive to economic growth, while mid-cap stocks are more sensitive to liquidity conditions [6][35][38]. Group 4: Performance of Different Strategies - The "lightweight" strategy, focusing on growth and inflation factors, has shown an annualized return of 7.7% with a volatility of 5.4% since 2016, outperforming traditional asset risk parity strategies [7][40][41]. - The "three-dimensional" strategy, incorporating M1, BCI, and PPI, has yielded an annualized return of 9.0% with a volatility of 7.8%, indicating a more diversified asset allocation [8][9]. - The "broad-spectrum" strategy, which includes multiple macro factors, has achieved an annualized return of 7.5% with a lower volatility of 4.0%, demonstrating a higher Sharpe ratio compared to simpler models [9][10].

Quantitative Models and Construction Methods 1. Model Name: PB-ROE-50 Combination - **Model Construction Idea**: This model combines Price-to-Book (PB) and Return on Equity (ROE) metrics to identify stocks with strong valuation and profitability characteristics[25] - **Model Construction Process**: - The PB-ROE-50 combination is constructed by selecting stocks based on their PB and ROE metrics, emphasizing stocks with favorable valuation and profitability profiles - The portfolio is rebalanced periodically to maintain the desired exposure to these factors[25] - **Model Evaluation**: The model demonstrates significant excess returns across multiple stock pools, indicating its effectiveness in capturing valuation and profitability signals[25] 2. Model Name: Block Trade Combination - **Model Construction Idea**: This model leverages the "high transaction, low volatility" principle to identify stocks with favorable post-trade performance based on block trade characteristics[31] - **Model Construction Process**: - Stocks are selected based on two key metrics: "block trade transaction amount ratio" and "6-day transaction amount volatility" - Stocks with higher transaction ratios and lower volatility are included in the portfolio - The portfolio is rebalanced monthly to capture updated signals[31] - **Model Evaluation**: The model effectively extracts excess information from block trades, yielding consistent positive returns relative to the benchmark[31] 3. Model Name: Private Placement Combination - **Model Construction Idea**: This model focuses on the event-driven effects of private placements, considering factors such as market value and timing of announcements[37] - **Model Construction Process**: - Stocks involved in private placements are selected based on the announcement date of shareholder meetings - Adjustments are made for market value factors, rebalancing cycles, and position control to optimize the portfolio[37] - **Model Evaluation**: The model captures the investment opportunities associated with private placements, delivering notable excess returns over the benchmark[37] --- Model Backtesting Results 1. PB-ROE-50 Combination - **Excess Return**: - CSI 500: 0.45% (weekly), 2.64% (YTD)[26] - CSI 800: 1.87% (weekly), 3.86% (YTD)[26] - All Market: 3.35% (weekly), 5.86% (YTD)[26] - **Absolute Return**: - CSI 500: 2.38% (weekly), 3.30% (YTD)[26] - CSI 800: 2.20% (weekly), 2.83% (YTD)[26] - All Market: 4.72% (weekly), 7.74% (YTD)[26] 2. Block Trade Combination - **Excess Return**: 0.41% (weekly), 23.89% (YTD)[32] - **Absolute Return**: 1.89% (weekly), 26.09% (YTD)[32] 3. Private Placement Combination - **Excess Return**: 1.97% (weekly), 6.08% (YTD)[38] - **Absolute Return**: 3.48% (YTD)[38] --- Quantitative Factors and Construction Methods 1. Factor Name: Total Asset Growth Rate - **Factor Construction Idea**: Measures the growth in total assets to capture expansion potential[12] - **Factor Construction Process**: - Calculated as the percentage change in total assets over a specified period - Adjusted for industry and market capitalization effects to isolate the factor signal[12] - **Factor Evaluation**: Demonstrates strong positive returns across multiple stock pools, indicating its effectiveness in identifying growth opportunities[12][18] 2. Factor Name: Single-Quarter ROA - **Factor Construction Idea**: Reflects the profitability of assets on a quarterly basis[12] - **Factor Construction Process**: - Calculated as net income divided by total assets for a single quarter - Adjusted for industry and market capitalization effects to enhance signal clarity[12] - **Factor Evaluation**: Consistently positive performance across stock pools, highlighting its robustness in capturing profitability signals[12][18] 3. Factor Name: Single-Quarter Revenue Growth Rate - **Factor Construction Idea**: Tracks the growth in revenue on a quarterly basis to identify companies with improving top-line performance[12] - **Factor Construction Process**: - Calculated as the percentage change in revenue compared to the same quarter in the previous year - Adjusted for industry and market capitalization effects to ensure comparability[12] - **Factor Evaluation**: Strong positive returns in multiple stock pools, validating its ability to capture growth momentum[12][18] --- Factor Backtesting Results 1. Total Asset Growth Rate - **Excess Return**: - CSI 300: 2.23% (weekly)[12] - CSI 500: 1.26% (weekly)[14] - Liquidity 1500: 2.67% (weekly)[18] 2. Single-Quarter ROA - **Excess Return**: - CSI 300: 1.58% (weekly)[12] - CSI 500: -0.44% (weekly)[15] - Liquidity 1500: 0.88% (weekly)[19] 3. Single-Quarter Revenue Growth Rate - **Excess Return**: - CSI 300: 1.78% (weekly)[12] - CSI 500: 0.58% (weekly)[15] - Liquidity 1500: 2.13% (weekly)[19]

Quantitative Models and Construction Methods - **Model Name**: Neutral Constraint Maximum Factor Exposure Portfolio **Model Construction Idea**: The model aims to maximize the exposure of a target factor in the portfolio while maintaining neutrality in industry and style exposures relative to the benchmark index[59][60][61] **Model Construction Process**: 1. Objective Function: Maximize the portfolio's exposure to the target factor $Max \ w^{\prime} X_{target}$ 2. Constraints: - Industry neutrality: $(w-w_{b})^{\prime} X_{ind}=0$ - Style neutrality (size, valuation, growth): $(w-w_{b})^{\prime} X_{Beta}=0$ - Stock weight deviation from benchmark: $|w-w_{b}|\leq1\%$ - No short selling: $w\geq0$ - Full investment: $w^{\prime} 1=1$ - Stocks must belong to the benchmark: $w^{\prime} B=1$ 3. Factor neutralization: Before constructing the portfolio, factors are neutralized to remove correlations with industry and style factors, and all factor directions are adjusted to be positive[59][60][61] **Model Evaluation**: The model effectively balances factor exposure maximization with risk control through constraints, ensuring robustness in various market conditions[59][60][61] --- Model Backtesting Results - **Neutral Constraint Maximum Factor Exposure Portfolio** - **CSI 300 Enhanced Portfolio**: Weekly excess return 0.35%, monthly excess return 0.33%, annual excess return 0.40%[56] - **CSI 500 Enhanced Portfolio**: Weekly excess return -0.52%, monthly excess return 1.34%, annual excess return -0.05%[56] - **CSI 800 Enhanced Portfolio**: Weekly excess return 0.29%, monthly excess return 1.59%, annual excess return 0.74%[56] - **CSI 1000 Enhanced Portfolio**: Weekly excess return 0.25%, monthly excess return 2.83%, annual excess return 15.68%[57] - **CSI 300 ESG Enhanced Portfolio**: Weekly excess return 0.14%, monthly excess return 0.62%, annual excess return 5.94%[57] --- Quantitative Factors and Construction Methods - **Factor Name**: Beta Factor **Factor Construction Idea**: Measures the sensitivity of a stock's returns to the market's returns, capturing risk preferences in the market[15][16] **Factor Construction Process**: - Calculate the stock's daily returns over the past 252 trading days - Perform an exponentially weighted regression of the stock's returns against the market index (CSI All Share Index) with a half-life of 63 days - Use the regression coefficient as the Beta value[15][16] **Factor Evaluation**: The Beta factor effectively captures market risk preferences, as evidenced by its strong performance in high-risk environments[15][16] - **Factor Name**: Size Factor **Factor Construction Idea**: Captures the size effect, where smaller-cap stocks tend to outperform larger-cap stocks[15][16] **Factor Construction Process**: - Compute the natural logarithm of the total market capitalization of each stock[15][16] **Factor Evaluation**: The size factor consistently demonstrates the small-cap effect, particularly in high-volatility markets[15][16] - **Factor Name**: Momentum Factor **Factor Construction Idea**: Identifies stocks with strong past performance, assuming trends persist in the short term[15][16] **Factor Construction Process**: - Calculate cumulative returns over the past 504 trading days, excluding the most recent 21 days - Apply an exponentially weighted average with a half-life of 126 days to the return series[15][16] **Factor Evaluation**: The momentum factor is effective in trending markets but may underperform during reversals[15][16] --- Factor Backtesting Results - **Beta Factor**: Weekly long-short return 2.61%, monthly long-short return -1.82%[18] - **Size Factor**: Weekly long-short return -2.11%, monthly long-short return -8.87%[18] - **Momentum Factor**: Weekly long-short return 0.58%, monthly long-short return -1.85%[18] --- Stock Selection Factors and Performance - **Factor Name**: Single Quarter ROE **Factor Construction Idea**: Measures profitability by comparing net income to shareholder equity for a single quarter[20][21] **Factor Construction Process**: - Calculate the ratio of net income attributable to shareholders to total shareholder equity for the most recent quarter[20][21] **Factor Backtesting Results**: - CSI 300: Weekly excess return 0.72%, monthly excess return 1.90%, annual excess return 5.43%[23] - CSI 500: Weekly excess return 0.85%, monthly excess return 0.91%, annual excess return 5.90%[29] - CSI 800: Weekly excess return 1.02%, monthly excess return 2.06%, annual excess return 3.95%[32] - CSI 1000: Weekly excess return 1.09%, monthly excess return 2.44%, annual excess return -3.47%[36] - **Factor Name**: Single Quarter EP **Factor Construction Idea**: Measures earnings yield by comparing net income to market capitalization for a single quarter[20][21] **Factor Construction Process**: - Calculate the ratio of net income attributable to shareholders to total market capitalization for the most recent quarter[20][21] **Factor Backtesting Results**: - CSI 300: Weekly excess return 0.89%, monthly excess return 1.65%, annual excess return 0.86%[23] - CSI 500: Weekly excess return 0.50%, monthly excess return 1.87%, annual excess return -4.22%[29] - CSI 800: Weekly excess return 1.06%, monthly excess return 2.04%, annual excess return -1.54%[32] - CSI 1000: Weekly excess return 0.38%, monthly excess return 1.69%, annual excess return -5.99%[36] - **Factor Name**: 20-Day Reversal **Factor Construction Idea**: Captures short-term mean reversion by focusing on stocks with recent underperformance[20][21] **Factor Construction Process**: - Calculate cumulative returns over the past 20 trading days[20][21] **Factor Backtesting Results**: - CSI 300: Weekly excess return 0.11%, monthly excess return -0.15%, annual excess return 8.90%[23] - CSI 500: Weekly excess return 0.80%, monthly excess return 1.57%, annual excess return 3.33%[29] - CSI 800: Weekly excess return 0.39%, monthly excess return 0.59%, annual excess return 8.27%[32] - CSI 1000: Weekly excess return 0.64%, monthly excess return 1.38%, annual excess return -6.69%[36]

Group 1 - The core viewpoint of the article highlights the performance of various market factors during the week of May 19 to May 23, 2025, indicating that momentum and growth factors yielded positive returns while liquidity, beta, and size factors experienced significant negative returns [2][3]. - In the CSI 300 stock pool, the best-performing factors included net profit discontinuity (1.30%), 5-day index moving average of trading volume (1.15%), and total asset gross profit margin TTM (1.02%) [3]. - In the CSI 500 stock pool, the top-performing factors were gross profit margin TTM (1.65%), single-quarter ROA (1.40%), and single-quarter total asset gross profit margin (1.26%) [3]. - The liquidity 1500 stock pool showed that the best-performing factors were 5-day average turnover rate (0.45%), 5-minute return skewness (0.36%), and downward volatility ratio (0.33%) [3]. Group 2 - The net asset growth rate factor performed well across various industries, while the net profit growth rate factor excelled in the steel industry [4]. - The earnings per share factor showed strong performance in the beauty and personal care industry, and the operating profit TTM factor performed well in the coal industry [4]. - The 5-day momentum factor exhibited significant momentum effects in the comprehensive industry, while reversal effects were notable in the oil and petrochemical, and food and beverage industries [4]. Group 3 - The PB-ROE-50 combination achieved significant excess returns in the CSI 500 stock pool, with an excess return of 1.15% [6]. - The public fund research stock selection strategy and private fund research tracking strategy both generated positive excess returns, with the public fund strategy outperforming the CSI 800 by 0.54% and the private fund strategy outperforming by 2.61% [7]. - The block trading combination experienced a decline in excess returns relative to the CSI All Index, with an excess return of -0.61% [8]. - The targeted issuance combination achieved excess returns relative to the CSI All Index, with an excess return of 0.12% [9].