Quantitative Models and Factors Summary Quantitative Models and Construction Methods - **Model Name**: Rotation Model - **Model Construction Idea**: The model uses relative valuation and momentum indicators to determine allocation between micro-cap stocks and "Mao Index" (a proxy for large-cap stocks) to capture style rotation opportunities[18][25] - **Model Construction Process**: 1. **Rotation Indicators**: - Calculate the relative net value of micro-cap stocks to the Mao Index - Compare the relative net value to its 243-day moving average. If above, favor micro-cap stocks; otherwise, favor the Mao Index - Use the 20-day closing price slope of both indices. If one slope is positive and the other is negative, allocate to the index with a positive slope[18][25] 2. **Timing Indicators**: - Use the 10-year government bond yield (threshold: 0.3%) and micro-cap stock volatility crowding degree (threshold: 0.55). If either indicator hits its threshold, issue a closing signal[25] - **Model Evaluation**: The model currently signals a balanced allocation between micro-cap stocks and the Mao Index, with no systemic risk triggers observed in the medium term[18][19] Quantitative Factors and Construction Methods - **Factor Name**: Growth Factor - **Factor Construction Idea**: Measures the growth potential of stocks based on financial metrics like revenue and profit growth[65] - **Factor Construction Process**: - Key metrics include: - **OperatingIncome_SQ_Chg1Y**: Year-over-year growth in quarterly operating income - **Revenues_SQ_Chg1Y**: Year-over-year growth in quarterly revenue - **ROE_FTTM**: Forward 12-month return on equity based on consensus estimates[65] - **Factor Evaluation**: The growth factor performed well in the past week, driven by market sentiment favoring growth-oriented stocks[54][56] - **Factor Name**: Consensus Expectation Factor - **Factor Construction Idea**: Captures market sentiment and analyst expectations through forward-looking metrics[65] - **Factor Construction Process**: - Key metrics include: - **ROE_FTTM_Chg3M**: 3-month change in forward 12-month ROE estimates - **TargetReturn_180D**: Expected return based on consensus target price over the next 180 days - **Volume_Mean_20D_240D**: Ratio of 20-day average trading volume to 240-day average trading volume[65] - **Factor Evaluation**: This factor exhibited strong performance last week, reflecting improved market sentiment and positive analyst revisions[54][56] - **Factor Name**: Volatility Factor - **Factor Construction Idea**: Measures the risk and defensive characteristics of stocks based on historical price volatility[65] - **Factor Construction Process**: - Key metrics include: - **IV_CAPM**: Residual volatility from the CAPM model - **IV_FF**: Residual volatility from the Fama-French three-factor model - **Volatility_60D**: Standard deviation of 60-day returns[65] - **Factor Evaluation**: The volatility factor weakened last week due to a decline in risk-averse sentiment as geopolitical tensions eased[54][56] - **Factor Name**: Reversal Factor - **Factor Construction Idea**: Exploits mean-reversion tendencies in stock prices over different time horizons[65] - **Factor Construction Process**: - Key metrics include: - **Price_Chg60D**: 60-day return - **Price_Chg120D**: 120-day return[65] - **Factor Evaluation**: The reversal factor underperformed last week, reflecting a market preference for momentum and growth[54][56] - **Factor Name**: Convertible Bond Selection Factors - **Factor Construction Idea**: Constructs factors based on the relationship between convertible bonds and their underlying stocks, as well as valuation metrics[59] - **Factor Construction Process**: - Key metrics include: - **Equity Consensus Expectation**: Derived from the underlying stock's consensus estimates - **Equity Growth**: Based on the growth metrics of the underlying stock - **Equity Financial Quality**: Evaluates the financial health of the underlying stock - **Equity Valuation**: Assesses the valuation of the underlying stock - **Convertible Bond Valuation**: Uses metrics like parity and premium rate[59][65] - **Factor Evaluation**: The equity consensus expectation factor achieved the highest IC mean among convertible bond factors last week[59][63] Backtest Results of Models and Factors - **Rotation Model**: - Relative net value of micro-cap stocks to Mao Index: 2.45 (above the 243-day moving average of 2.00)[18][25] - 20-day closing price slope: Micro-cap stocks -0.37%, Mao Index -0.17%[18][25] - Volatility crowding degree: 12.43% (below the risk threshold of 55%)[18][25] - 10-year government bond yield: 0.61% (below the risk threshold of 0.3%)[18][25] - **Factor Backtest Results (IC Mean)**: - **Consensus Expectation**: 7.37% (All A-shares), 0.08% (CSI 300), 6.25% (CSI 500), 2.85% (CSI 1000)[56] - **Growth**: 0.90% (All A-shares), 4.95% (CSI 300), 3.53% (CSI 500), 4.77% (CSI 1000)[56] - **Volatility**: -4.38% (All A-shares), -6.04% (CSI 300), -10.10% (CSI 500), -8.15% (CSI 1000)[56] - **Reversal**: -12.58% (All A-shares), -4.57% (CSI 300), -2.55% (CSI 500), -8.97% (CSI 1000)[56] - **Convertible Bond Factors (IC Mean)**: - Equity Consensus Expectation: Highest IC mean among all convertible bond factors[59][63]

Quantitative Models and Construction Methods - **Model Name**: Three-dimensional Timing Framework **Model Construction Idea**: The model integrates liquidity, divergence, and prosperity indicators to assess market timing[6][13][15] **Model Construction Process**: 1. Liquidity Index: Measures market liquidity trends[24] 2. Divergence Index: Captures market disagreement levels[19] 3. Prosperity Index: Reflects economic activity and market sentiment[22] These three dimensions are combined to form a comprehensive timing framework[13][15] **Model Evaluation**: The framework indicates a downward market trend with limited short-term rebound potential[6][13] - **Model Name**: All-weather Strategy **Model Construction Idea**: Focuses on risk diversification and avoids reliance on predictions for stable returns[42][53] **Model Construction Process**: 1. Asset Selection: Diversified across equities, bonds, and commodities[55] 2. Risk Adjustment: Balances risk exposure through structured layers[46][48] 3. Structural Hedging: Implements cyclic hedging to smooth volatility[42][47][48] **Model Evaluation**: High-wave version achieves higher returns with moderate risk, while low-wave version prioritizes stability[53] - **Model Name**: Hotspot Trend ETF Strategy **Model Construction Idea**: Identifies ETFs with strong upward trends and high market attention[29][32] **Model Construction Process**: 1. Select ETFs with simultaneous upward trends in highest and lowest prices[29] 2. Construct support-resistance factors based on 20-day regression slopes[29] 3. Choose top ETFs with the highest turnover ratios in the past 5 and 20 days[29] **Model Evaluation**: The strategy outperformed the CSI 300 index with a 56.47% return since 2025[29][30] Model Backtesting Results - **Three-dimensional Timing Framework**: No specific numerical backtesting results provided - **All-weather Strategy**: - High-wave version: Annualized return 11.8%, max drawdown 3.6%, Sharpe ratio 1.9 (2025)[53] - Low-wave version: Annualized return 6.7%, max drawdown 2.0%, Sharpe ratio 2.4 (2025)[53] - 2026 YTD: High-wave return 1.8%, low-wave return 1.2%[53] - **Hotspot Trend ETF Strategy**: - Return since 2025: 56.47% - Excess return over CSI 300: 38.62%[29][30] Quantitative Factors and Construction Methods - **Factor Name**: Volatility Factor **Factor Construction Idea**: Captures stocks with high price fluctuations[56] **Factor Construction Process**: Measures weekly returns of high-volatility stocks[56] **Factor Evaluation**: Positive weekly return of 1.95%, indicating market preference for high-volatility stocks[56][57] - **Factor Name**: Momentum Factor **Factor Construction Idea**: Identifies stocks with strong upward price trends[56] **Factor Construction Process**: Calculates weekly returns of high-momentum stocks[56] **Factor Evaluation**: Positive weekly return of 1.58%, reflecting market interest in momentum stocks[56][57] - **Factor Name**: Leverage Factor **Factor Construction Idea**: Targets stocks with high financial leverage[56] **Factor Construction Process**: Measures weekly returns of high-leverage stocks[56] **Factor Evaluation**: Positive weekly return of 0.96%, showing market favor for leveraged stocks[56][57] - **Factor Name**: Twelve-month Residual Momentum **Factor Construction Idea**: Tracks residual momentum over a 12-month period[61] **Factor Construction Process**: $ specific\_mom12 = residual\_momentum\_12months $ Measures excess returns of stocks with strong residual momentum[61][62] **Factor Evaluation**: Weekly excess return of 0.87%, monthly excess return of 0.46%[61][62] - **Factor Name**: 1-year-1-month Return Factor **Factor Construction Idea**: Compares returns between 1-year and 1-month periods[61] **Factor Construction Process**: $ mom\_1y\_1m = (return\_1year - return\_1month) $ Calculates excess returns based on the difference between long-term and short-term returns[61][62] **Factor Evaluation**: Weekly excess return of 0.79%, monthly excess return of -0.03%[61][62] Factor Backtesting Results - **Volatility Factor**: Weekly return 1.95%[56][57] - **Momentum Factor**: Weekly return 1.58%[56][57] - **Leverage Factor**: Weekly return 0.96%[56][57] - **Twelve-month Residual Momentum**: Weekly excess return 0.87%, monthly excess return 0.46%[61][62] - **1-year-1-month Return Factor**: Weekly excess return 0.79%, monthly excess return -0.03%[61][62]

Quantitative Models and Construction Methods - **Model Name**: All-Weather Strategy **Model Construction Idea**: The strategy aims to achieve stable returns by avoiding reliance on predictions, leveraging diversified risk allocation principles[44][53] **Model Construction Process**: 1. **Asset Selection**: Diversify across equities, bonds, and commodities 2. **Risk Adjustment**: Balance risk exposure through structured layers 3. **Structural Hedging**: Implement cyclic hedging to smooth volatility - High-volatility version: Four-layer structure focusing on equity, bond, and gold risk parity - Low-volatility version: Five-layer structure emphasizing risk budgeting **Model Evaluation**: The strategy effectively balances risk and return, achieving stable absolute returns without relying on leverage or macroeconomic assumptions[44][53] - **Model Name**: Hotspot Trend ETF Strategy **Model Construction Idea**: Identify ETFs with strong short-term market attention and construct a risk-parity portfolio[32] **Model Construction Process**: 1. Select ETFs with both highest and lowest price trends in the past 20 days 2. Construct support-resistance factors based on the steepness of regression coefficients of the highest and lowest prices 3. Choose the top 10 ETFs with the highest turnover ratio (5-day/20-day) from the long factor group 4. Construct a risk-parity portfolio using these ETFs **Model Evaluation**: The strategy demonstrates strong excess returns over the benchmark, indicating its effectiveness in capturing market trends[32][35] - **Model Name**: Capital Flow Resonance Strategy **Model Construction Idea**: Combine financing and large-order capital flow factors to identify industries with capital flow resonance[40] **Model Construction Process**: 1. Define financing capital flow factor: Neutralize the financing net buy-sell data by market capitalization and calculate the 50-day average two-week change rate 2. Define large-order capital flow factor: Neutralize the industry’s one-year transaction volume and calculate the 10-day average net inflow ranking 3. Combine the two factors, excluding extreme industries and large financial sectors 4. Construct a weekly rebalancing strategy based on the combined factor scores **Model Evaluation**: The strategy achieves stable positive excess returns with reduced drawdowns compared to other capital flow strategies[40][42] Model Backtesting Results - **All-Weather Strategy**: - High-volatility version: Annualized return 11.8%, maximum drawdown 3.6%, Sharpe ratio 1.9 (2025 data)[53] - Low-volatility version: Annualized return 6.7%, maximum drawdown 2.0%, Sharpe ratio 2.4 (2025 data)[53] - 2026 YTD returns: High-volatility version 1.9%, low-volatility version 1.1%[53] - **Hotspot Trend ETF Strategy**: - 2025 cumulative return: 58.34% - Excess return over CSI 300 Index: 38.80%[32][35] - **Capital Flow Resonance Strategy**: - Annualized excess return since 2018: 14.3% - Information ratio (IR): 1.4 - Weekly absolute return: -2.53%, excess return: 1.88% (latest week)[40][42] Quantitative Factors and Construction Methods - **Factor Name**: Return Standard Deviation (1 Month) **Factor Construction Idea**: Measure the standard deviation of returns over the past month to capture volatility trends[61] **Factor Construction Process**: 1. Calculate daily returns over the past month 2. Compute the standard deviation of these returns **Factor Evaluation**: Demonstrates strong stock selection ability with consistent positive excess returns[61] - **Factor Name**: Average Turnover Rate (63 Days) **Factor Construction Idea**: Use the natural logarithm of the average turnover rate over the past 63 trading days to assess liquidity trends[61] **Factor Construction Process**: 1. Calculate the daily turnover rate for the past 63 trading days 2. Compute the natural logarithm of the average turnover rate **Factor Evaluation**: Exhibits robust performance in identifying high-liquidity stocks[61] - **Factor Name**: Consensus Forecast Net Profit Change (FY1) **Factor Construction Idea**: Measure the change in consensus forecast net profit (FY1) over different time horizons to capture earnings revisions[63] **Factor Construction Process**: 1. Calculate the difference between the current consensus forecast net profit (FY1) and the forecast from 1/3 months ago 2. Normalize the change by dividing it by the absolute value of the forecast from 1/3 months ago **Factor Evaluation**: Performs well in small-cap indices, reflecting market sensitivity to earnings revisions[63] Factor Backtesting Results - **Return Standard Deviation (1 Month)**: Weekly excess return 1.27%, monthly excess return 1.14%[62] - **Average Turnover Rate (63 Days)**: Weekly excess return 1.26%, monthly excess return 0.83%[62] - **Consensus Forecast Net Profit Change (FY1)**: - CSI 300: 28.03% (3-month horizon) - CSI 500: 16.98% (1-month horizon) - CSI 800: 26.83% (3-month horizon) - CSI 1000: 15.94% (3-month horizon)[63][64]

Core Viewpoint - The article discusses the transformative changes in China's private equity industry by 2025, highlighting the rise of AI-driven quantitative strategies, the growth of private equity scale to over 22 trillion yuan, accelerated overseas expansion, and a shift towards a diversified industry landscape [1]. Group 1: Personal Growth and Career Choices - The author shares a non-linear academic journey through three majors, which ultimately laid a unique foundation for a career in quantitative investing [3]. - The author emphasizes the importance of interdisciplinary knowledge, combining finance, statistics, programming, sociology, and psychology to understand market behaviors [9]. Group 2: Quantitative Methodology and Features of Zhongou Ruibo - Zhongou Ruibo has developed a systematic quantitative research strategy covering stocks, stock index futures, commodity futures, and government bond futures, enabling the capture of investment opportunities across markets and cycles [16]. - The stock model includes a rich factor library with hundreds of underlying factors, with 10% of factors being replaced or iterated annually to adapt to market changes [17]. - The stock index futures model focuses on four major index futures, with 20+ strategies, 80% of which are trend-following strategies [18]. Group 3: Team Building and Talent Development - The company seeks quantitative newcomers with solid academic training in statistics, finance, and programming, as well as a scientific research spirit [32]. - A practical training system is being developed, focusing on real investment needs to enhance the applicability of research outcomes [34]. - The author advises aspiring quantitative researchers to build strong foundations in finance, statistics, and programming, while also learning to utilize AI tools effectively [35][38]. Group 4: Performance and Risk Management - Zhongou Ruibo offers three main product types: a stock CTA composite product, a multi-strategy CTA fund, and a stock quantitative long strategy, emphasizing diversified investment to enhance resilience in extreme market conditions [28][29][30].

本订阅号中所涉及的证券研究信息由光大证券研究所编写，仅面向光大证券专业投资者客户，用作新媒体形势下研究 信息和研究观点的沟通交流。非光大证券专业投资者客户，请勿订阅、接收或使用本订阅号中的任何信息。本订阅号 难以设置访问权限，若给您造成不便，敬请谅解。光大证券研究所不会因关注、收到或阅读本订阅号推送内容而视相 关人员为光大证券的客户。 报告摘要 量化市场跟踪 大类因子表现： 点击注册小程序 查看完整报告 特别申明： PB-ROE-50组合跟踪： 本周全市场股票池中，残差波动率因子、市值因子分别获取正收益0.40%、0.40%，市场表现为大市值风 格；流动性因子、估值因子、Beta因子分别获取负收益-0.43%、-0.39%、-0.36%；其余风格因子表现一 般。 单因子表现： 沪深300股票池中，本周表现较好的因子有总资产增长率 (2.17%)、单季度总资产毛利率 (1.11%)、单季度 ROE (1.11%)。表现较差的因子有早盘收益因子 (-4.08%)、5分钟收益率偏度 (-3.62%)、动量弹簧因子 (-3.52%)。 中证500股票池中，本周表现较好的因子有总资产增长率 (2.99%)、单季度营业收 ...

- The report tracks the performance of single factors in different stock pools, including the CSI 300, CSI 500, and liquidity 1500 pools. Positive-performing factors in the CSI 300 pool include total asset growth rate (2.17%), single-quarter total asset gross margin (1.11%), and single-quarter ROE (1.11%). Negative-performing factors include morning return factor (-4.08%), 5-minute return skewness (-3.62%), and momentum spring factor (-3.52%) [12][13] - In the CSI 500 stock pool, factors with strong performance include total asset growth rate (2.99%), single-quarter operating revenue YoY growth rate (2.90%), and EPTTM percentile (2.83%). Poor-performing factors include 5-minute return skewness (-0.57%), 5-day reversal (-0.44%), and turnover rate relative volatility (-0.41%) [14][15] - For the liquidity 1500 stock pool, top-performing factors are single-quarter operating revenue YoY growth rate (2.33%), total asset growth rate (2.20%), and single-quarter ROE (2.05%). Factors with weaker performance include morning return factor (-1.09%), PE TTM reciprocal (-1.04%), and PS TTM reciprocal (-0.95%) [16][17] - The report highlights the performance of broad categories of factors across the entire market stock pool. Residual volatility factor and market capitalization factor achieved positive returns of 0.40% each, while liquidity factor (-0.43%), valuation factor (-0.39%), and Beta factor (-0.36%) underperformed [18][20] - Industry-specific factor performance is analyzed, showing that basic fundamental factors like net asset per share and operating profit per share TTM performed consistently well in the petroleum and petrochemical and transportation industries. Valuation factors such as BP and EP also showed consistent positive returns in the petroleum and petrochemical sector. Residual volatility and liquidity factors performed well in the construction and decoration industry [21][22] - The PB-ROE-50 portfolio achieved positive excess returns in the CSI 500 stock pool (0.53%) but negative excess returns in the CSI 800 (-1.35%) and the entire market stock pool (-0.76%) [23][24] - The institutional research portfolios, including public fund research stock selection and private fund research tracking strategies, both recorded negative excess returns relative to the CSI 800 index. Public fund research stock selection strategy achieved -3.57%, while private fund research tracking strategy achieved -2.62% [25][26] - The block trading portfolio, constructed based on "high transaction, low volatility" principles, recorded negative excess returns of -1.88% relative to the CSI All Index [29][30] - The directed issuance portfolio, constructed around event-driven strategies tied to shareholder meeting announcement dates, achieved positive excess returns of 0.68% relative to the CSI All Index [35][36]

Core Viewpoint - The article provides a comprehensive 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 the effectiveness of specific investment strategies [4][5][6][8][9][10][11]. Factor Performance - In the overall market, valuation, profitability, leverage, and market capitalization factors achieved positive returns of 0.43%, 0.35%, 0.27%, and 0.26% respectively, indicating a large-cap style market [4]. - Conversely, momentum and growth factors recorded negative returns of -0.66% and -0.28%, suggesting a reversal effect in the market [4]. Single Factor Performance - In the CSI 300 stock pool, the best-performing factors included operating cash flow ratio (1.68%), 5-day volume moving average (1.19%), and total asset gross margin TTM (0.93%). The worst performers were small net inflow (-2.99%), 5-day reversal (-2.42%), and momentum-adjusted small net (-1.62%) [5]. - In the CSI 500 stock pool, the top factors were downside volatility ratio (3.30%), price-to-earnings ratio (3.24%), and inverse of TTM price-to-earnings ratio (3.06%). The underperformers included single-quarter total asset gross margin (-0.51%), small net inflow (-0.33%), and single-quarter ROA (-0.31%) [5]. - In the liquidity 1500 stock pool, the leading factors were inverse of TTM price-to-earnings ratio (1.56%), price-to-book ratio (1.33%), and single-quarter EPS (0.65%). The lagging factors were TTM gross margin (-1.94%), TTM total asset gross margin (-1.83%), and standardized expected external income (-1.58%) [5]. Industry Factor Performance - Fundamental factors showed varied performance across industries, with net asset per share and operating profit TTM factors yielding consistent positive returns in coal and diversified industries. Valuation factors like BP and EP showed significant positive returns across most industries [6]. - In the real estate sector, residual volatility and liquidity factors exhibited positive returns, while large-cap styles were notably strong in utilities, electrical equipment, and construction decoration industries [6]. Investment Strategy Performance - The PB-ROE-50 combination in the CSI 500 stock pool achieved a positive excess return of 0.77%, while the CSI 800 stock pool recorded a negative excess return of -1.15%, and the overall market stock pool had a negative excess return of -1.79% [8]. - The public fund research stock selection strategy generated a positive excess return of 0.26% relative to the CSI 800, while the private fund research tracking strategy had a negative excess return of -2.32% [9]. - The block trading combination achieved a positive excess return of 0.92% relative to the CSI All Index [10]. - The targeted issuance combination recorded a negative excess return of -0.87% relative to the CSI All Index [11].

- The report tracks the performance of various factors in different stock pools, including the CSI 300, CSI 500, and Liquidity 1500 stock pools[1][2][3] - In the CSI 300 stock pool, the best-performing factors this week were Operating Cash Flow Ratio (1.68%), 5-day Exponential Moving Average of Volume (1.19%), and Total Asset Gross Profit Margin TTM (0.93%)[12] - In the CSI 500 stock pool, the best-performing factors this week were Downside Volatility Ratio (3.30%), PE Ratio Factor (3.24%), and PE Ratio TTM Reciprocal (3.06%)[14] - In the Liquidity 1500 stock pool, the best-performing factors this week were PE Ratio TTM Reciprocal (1.56%), PB Ratio Factor (1.33%), and Quarterly EPS (0.65%)[16] - The PB-ROE-50 portfolio achieved positive excess returns in the CSI 500 stock pool (0.77%) but negative excess returns in the CSI 800 stock pool (-1.15%) and the overall market stock pool (-1.79%)[23][24] - The institutional research portfolio tracking showed that the public research stock selection strategy achieved positive excess returns (0.26%) relative to the CSI 800, while the private research tracking strategy achieved negative excess returns (-2.32%)[25][26] - The block trading portfolio achieved positive excess returns (0.92%) relative to the CSI All Share Index[29][30] - The directed issuance portfolio achieved negative excess returns (-0.87%) relative to the CSI All Share Index[35][36]

Quantitative Models and Construction Methods 1. 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 terms of industry and style exposures relative to the benchmark index[59][60][62] - **Model Construction Process**: 1. The objective function is to maximize the portfolio's exposure to the target factor 2. Constraints include: - Industry neutrality: The portfolio's industry exposure relative to the benchmark index is controlled to be zero - Style neutrality: The portfolio's exposure to size, valuation, and growth factors relative to the benchmark index is controlled to be zero - Stock weight deviation: The weight of each stock in the portfolio relative to its weight in the benchmark index is limited to a maximum deviation of 1% - No short selling is allowed - All portfolio components must be constituents of the benchmark index - The sum of weights equals 1, ensuring the portfolio is fully invested 3. The optimization model is expressed as: $ \begin{array}{l} \mbox{\it Max}\quad\quad\quad w^{\prime}\;X_{target}\\ \mbox{\it s.t.}\quad\quad\quad(w-\;w_{b})^{\prime}X_{ind}=\;0\\ \mbox{\it(w-\;w_{b})}^{\prime}\;X_{Beta}=\;0\\ \mbox{\it|w-\;w_{b}|\leq1\%}\\ \mbox{\it w\geq0}\\ \mbox{\it w^{\prime}B=1}\\ \mbox{\it w^{\prime}1=1} \end{array} $ where $w$ represents the weight vector of individual stocks in the portfolio, $w_b$ represents the weight vector of individual stocks in the benchmark portfolio, $X_{target}$ is the factor loading matrix for the target factor, $X_{ind}$ is the industry exposure matrix, and $X_{Beta}$ is the factor loading matrix for style factors (size, valuation, growth)[59][60][62] 4. Before constructing the portfolio, factors are neutralized to remove their correlation with industry and style factors, and all factor directions are adjusted to be positive[61] - **Model Evaluation**: The model ensures that the portfolio remains neutral to industry and style exposures while maximizing the target factor exposure[62] --- Model Backtesting Results 1. Neutral Constraint Maximum Factor Exposure Portfolio - **CSI 300 Enhanced Portfolio**: - Weekly excess return: 0.15% - Monthly excess return: 1.96% - Annual excess return: 15.99% - Information ratio (IR): 2.33 (full sample period)[55][57][58] - **CSI 500 Enhanced Portfolio**: - Weekly excess return: 1.03% - Monthly excess return: -0.12% - Annual excess return: -10.53% - IR: 1.96 (full sample period)[55][57][58] - **CSI 800 Enhanced Portfolio**: - Weekly excess return: 0.56% - Monthly excess return: 1.07% - Annual excess return: 9.60% - IR: 2.12 (full sample period)[55][57][58] - **CSI 1000 Enhanced Portfolio**: - Weekly excess return: 0.97% - Monthly excess return: 1.16% - Annual excess return: 16.76% - IR: 2.91 (full sample period)[55][57][58] - **CSI 300 ESG Enhanced Portfolio**: - Weekly excess return: 0.72% - Monthly excess return: 0.32% - Annual excess return: 6.32% - IR: 1.77 (full sample period)[55][57][58] --- Quantitative Factors and Construction Methods 1. Factor Name: Valuation Factor (BP) - **Factor Construction Idea**: Captures the valuation level of stocks by comparing book value to market value[18][19] - **Factor Construction Process**: - Formula: $BP = \frac{\text{Book Value of Equity}}{\text{Market Value of Equity}}$[19] - **Factor Evaluation**: The factor performed well in recent periods, indicating that low valuation stocks outperformed high valuation stocks[18][19] 2. Factor Name: Profitability Factor (ETOP, CETOP) - **Factor Construction Idea**: Measures the profitability of stocks relative to their market value[18][19] - **Factor Construction Process**: - Formula: $ETOP = \frac{\text{Net Profit (TTM)}}{\text{Market Value}}$ - Formula: $CETOP = \frac{\text{Net Cash Flow from Operating Activities (TTM)}}{\text{Total Assets}}$ - Profitability factor = (ETOP + CETOP) / 2[19] - **Factor Evaluation**: The factor showed strong performance, with high profitability stocks outperforming low profitability stocks[18][19] 3. Factor Name: Momentum Factor (RSTR) - **Factor Construction Idea**: Captures the relative strength of stocks based on past returns[18][19] - **Factor Construction Process**: - Formula: $RSTR = \text{Cumulative Returns over the past 504 trading days (excluding the most recent 21 days)}$ - Returns are exponentially weighted with a half-life of 126 trading days[19] - **Factor Evaluation**: The factor demonstrated significant performance over the past month, indicating that high-momentum stocks outperformed[18][19] 4. Factor Name: Liquidity Factor (STOM, STOQ, STOA) - **Factor Construction Idea**: Measures the liquidity of stocks based on turnover rates over different time horizons[18][19] - **Factor Construction Process**: - Formula: $STOM = \text{Logarithm of the sum of turnover rates over the past 1 month}$ - Formula: $STOQ = \text{Average of STOM over the past 3 months}$ - Formula: $STOA = \text{Average of STOM over the past 12 months}$ - Liquidity factor = (STOM + STOQ + STOA) / 3[19] - **Factor Evaluation**: The factor underperformed recently, indicating that less liquid stocks outperformed more liquid stocks[18][19] --- Factor Backtesting Results 1. Valuation Factor (BP) - Weekly return: 1.05% (CSI 800), 1.55% (CSI 1000) - Monthly return: 0.71% (CSI 800), 0.62% (CSI 1000) - Annual return: 0.47% (CSI 800), 2.15% (CSI 1000)[33][36][44] 2. Profitability Factor (ETOP, CETOP) - Weekly return: 1.49% (CSI 800), 0.67% (CSI 1000) - Monthly return: 0.33% (CSI 800), 0.40% (CSI 1000) - Annual return: 7.49% (CSI 800), 5.64% (CSI 1000)[33][36][44] 3. Momentum Factor (RSTR) - Weekly return: 0.58% (CSI 800), -0.08% (CSI 1000) - Monthly return: 0.68% (CSI 800), -2.60% (CSI 1000) - Annual return: 1.73% (CSI 800), -8.11% (CSI 1000)[33][36][44] 4. Liquidity Factor (STOM, STOQ, STOA) - Weekly return: 0.96% (CSI 800), 0.86% (CSI 1000) - Monthly return: -1.02% (CSI 800), -0.25% (CSI 1000) - Annual return: -7.25% (CSI 800), -4.83% (CSI 1000)[33][36][44]

Group 1 - The core viewpoint of the article highlights the performance of various market factors, indicating that momentum and profitability factors yielded positive returns while the Beta factor showed negative returns [4] - In the CSI 300 stock pool, the best-performing factors included the inverse of TTM price-to-earnings ratio (2.97%) and the price-to-earnings factor (2.86%), while the worst-performing factors were the year-on-year growth rate of quarterly operating revenue (-2.54%) and quarterly ROA year-on-year (-1.79%) [5] - The CSI 500 stock pool showed strong performance in the inverse of TTM price-to-earnings ratio (3.46%) and the price-to-earnings factor (3.29%), with the worst-performing factors being the 5-day average turnover rate (-1.22%) and total asset growth rate (-0.97%) [5] Group 2 - The article notes that fundamental factors exhibited varied performance across industries, with net asset per share and TTM operating profit per share showing consistent positive returns in coal and comprehensive industries [6] - The PB-ROE-50 combination achieved positive excess returns in both the CSI 500 (0.78%) and CSI 800 (0.46%) stock pools, while the overall market stock pool experienced a negative excess return (-0.92%) [8] - The public fund research stock selection strategy generated positive excess returns relative to the CSI 800 (0.16%), while the private fund research tracking strategy underperformed with a negative excess return (-1.30%) [9]