- The "ERP and DRP standardized equal-weight calculation model" is used to compute A-share odds, which as of September end, declined to 0.2 standard deviations, indicating a neutral level[10] - The "macro victory rate scoring card model" synthesizes asset victory rates based on factors like credit and PMI pulses, which recently bottomed out, pushing A-share victory rates to 19%[10] - The "bond odds model" is constructed using the expected yield difference between long and short bonds, with recent bond odds retreating to -0.9 standard deviations, reflecting valuation risks for long bonds[11] - The "bond victory rate model" integrates credit and growth expansion data, showing a decline to -6%, indicating low victory rates[11] - The "AIAE indicator model" for US stocks is at 54%, its historical peak, corresponding to 2.4 standard deviations, signaling high pullback risks[15] - The "Federal Reserve liquidity index model" combines quantity and price dimensions, showing a tightening liquidity index at 20%, a medium-high level[15] Model Backtesting Results - ERP and DRP model: A-share odds at 0.2 standard deviations, victory rate at 19%[10] - Bond odds model: -0.9 standard deviations, victory rate at -6%[11] - AIAE indicator model: 54% historical peak, 2.4 standard deviations[15] - Federal Reserve liquidity index: 20% medium-high level[15] Factor Construction and Evaluation - Value factor: High odds (0.9 SD), medium trend (-0.3 SD), low crowding (-1.4 SD), comprehensive score 3, recommended for focus[19][22] - Small-cap factor: Medium odds (-0.2 SD), strong trend (1.6 SD), medium-low crowding (-0.5 SD), comprehensive score 2.2, configuration value improved[20][23] - Quality factor: High odds (1.4 SD), weak trend (-1.2 SD), medium-low crowding (-0.5 SD), comprehensive score 0.6, recommended for long-term attention[24][26] - Growth factor: Medium-high odds (0.8 SD), medium trend (0.1 SD), high crowding (1.0 SD), comprehensive score 0.1, recommended for standard allocation[27][28] Factor Backtesting Results - Value factor: Odds 0.9 SD, trend -0.3 SD, crowding -1.4 SD, score 3[19][22] - Small-cap factor: Odds -0.2 SD, trend 1.6 SD, crowding -0.5 SD, score 2.2[20][23] - Quality factor: Odds 1.4 SD, trend -1.2 SD, crowding -0.5 SD, score 0.6[24][26] - Growth factor: Odds 0.8 SD, trend 0.1 SD, crowding 1.0 SD, score 0.1[27][28] Strategy Construction and Evaluation - "Odds-enhanced strategy" allocates assets based on odds indicators under volatility constraints, achieving annualized returns of 6.6%-7.5% and maximum drawdowns of 2.4%-3.0% since 2011[39][41] - "Victory rate-enhanced strategy" uses macro victory rate scoring to allocate assets, achieving annualized returns of 6.3%-7.7% and maximum drawdowns of 2.3%-2.8% since 2011[42][44] - "Odds + victory rate strategy" combines risk budgets from both strategies, achieving annualized returns of 7.0%-7.6% and maximum drawdowns of 2.7%-2.8% since 2011[45][47] Strategy Backtesting Results - Odds-enhanced strategy: Annualized returns 6.6%-7.5%, max drawdowns 2.4%-3.0%[39][41] - Victory rate-enhanced strategy: Annualized returns 6.3%-7.7%, max drawdowns 2.3%-2.8%[42][44] - Odds + victory rate strategy: Annualized returns 7.0%-7.6%, max drawdowns 2.7%-2.8%[45][47]

Quantitative Models and Construction Valuation Spread-Based Style Factor Model - **Model Name**: Valuation Spread-Based Style Factor Model - **Model Construction Idea**: This model uses the valuation spread between long and short groups of a factor (e.g., BP) to construct a style factor's odds indicator. It emphasizes a "left-side" signal characteristic, meaning "buy when it drops significantly, sell when it rises significantly" [7][11] - **Model Construction Process**: 1. Select a factor and sort stocks by factor values into five groups (industry-neutral, grouping within each industry and then merging) to determine the long and short groups [11] 2. Calculate the median log(BP) for the long and short groups, where log transformation ensures BP follows a normal distribution [11] 3. Compute the raw valuation spread as the difference between the median log(BP) of the long group and the short group [11] 4. Standardize the raw valuation spread using a rolling six-year z-score [11] - **Model Evaluation**: The valuation spread demonstrates a certain degree of differentiation in predicting the future one-year returns of style factors. It is more suitable for identifying left-side signals [7] --- Quantitative Factors and Construction Dividend Yield Factor - **Factor Name**: Dividend Yield Factor - **Factor Construction Idea**: Represents the valuation spread of the dividend yield factor, indicating the relative attractiveness of value-oriented stocks [14] - **Factor Construction Process**: Constructed using the valuation spread methodology described above, with the dividend yield as the underlying factor [14] PB Factor - **Factor Name**: PB Factor - **Factor Construction Idea**: Represents the valuation spread of the price-to-book ratio (PB), reflecting the relative valuation of value-oriented stocks [14] - **Factor Construction Process**: Constructed using the valuation spread methodology described above, with PB as the underlying factor [14] PE Factor - **Factor Name**: PE Factor - **Factor Construction Idea**: Represents the valuation spread of the price-to-earnings ratio (PE), indicating the relative valuation of value-oriented stocks [14] - **Factor Construction Process**: Constructed using the valuation spread methodology described above, with PE as the underlying factor [14] Quality Factor - **Factor Name**: Quality Factor - **Factor Construction Idea**: Represents the valuation spread of the ROE factor, typically associated with "core assets" like high-quality companies [20] - **Factor Construction Process**: Constructed using the valuation spread methodology described above, with ROE as the underlying factor [20] Low Volatility Factor - **Factor Name**: Low Volatility Factor - **Factor Construction Idea**: Represents the valuation spread of low-volatility stocks, often linked to "stable assets" [20] - **Factor Construction Process**: Constructed using the valuation spread methodology described above, with volatility as the underlying factor [20] Momentum Factor - **Factor Name**: Momentum Factor - **Factor Construction Idea**: Represents the valuation spread of momentum stocks, often associated with stocks heavily held by public funds [20] - **Factor Construction Process**: Constructed using the valuation spread methodology described above, with momentum as the underlying factor [20] --- Backtesting Results of Factors Valuation Spread-Based Style Factor Model - **Odds Indicator**: Demonstrates differentiation in predicting future one-year returns of style factors, with higher odds indicating better opportunities [7][8] Dividend Yield Factor - **Odds**: 0.47 standard deviations, categorized as medium odds [14] PB Factor - **Odds**: 0.63 standard deviations, categorized as medium-high odds [14] PE Factor - **Odds**: 0.82 standard deviations, categorized as medium-high odds [14] Quality Factor - **Odds**: 1.17 standard deviations, categorized as high odds [20] Low Volatility Factor - **Odds**: 1.75 standard deviations, categorized as very high odds [20] Momentum Factor - **Odds**: -1.36 standard deviations, categorized as low odds [20]

Quantitative Models and Construction 1. Model Name: Odds + Win Rate Strategy - **Model Construction Idea**: This strategy combines the risk budget of the odds-based strategy and the win-rate-based strategy to create a comprehensive scoring system for asset allocation[3][48][54] - **Model Construction Process**: 1. The odds-based strategy allocates more to high-odds assets and less to low-odds assets under a target volatility constraint[48] 2. The win-rate-based strategy derives macro win-rate scores from five factors: monetary, credit, growth, inflation, and overseas, and allocates accordingly[51] 3. The combined strategy sums the risk budgets of the two strategies to form a unified allocation model[54] - **Model Evaluation**: The model demonstrates stable performance with low drawdowns and consistent returns over different time periods[54] 2. Model Name: Industry Rotation Strategy - **Model Construction Idea**: This strategy evaluates industries based on three dimensions: momentum/trend, turnover/volatility/beta (crowding), and IR (information ratio) over the past 12 months[43] - **Model Construction Process**: 1. Momentum and trend are measured using the IR of industries over the past 12 months[43] 2. Crowding is assessed using turnover ratio, volatility ratio, and beta ratio[43] 3. The strategy ranks industries based on these metrics and allocates to those with strong trends, low crowding, and high IR[43] - **Model Evaluation**: The strategy has shown strong excess returns and low tracking errors, making it a robust framework for industry allocation[43] --- Model Backtesting Results 1. Odds + Win Rate Strategy - **Annualized Return**: - 2011 onwards: 7.0% - 2014 onwards: 7.6% - 2019 onwards: 7.2%[54] - **Maximum Drawdown**: - 2011 onwards: 2.8% - 2014 onwards: 2.7% - 2019 onwards: 2.8%[54] - **Sharpe Ratio**: - 2011 onwards: 2.86 - 2014 onwards: 3.26 - 2019 onwards: 2.85[56] 2. Industry Rotation Strategy - **Excess Return**: - 2011 onwards: 13.1% - 2014 onwards: 13.0% - 2019 onwards: 10.8%[44] - **Tracking Error**: - 2011 onwards: 11.0% - 2014 onwards: 12.0% - 2019 onwards: 10.7%[44] - **IR**: - 2011 onwards: 1.18 - 2014 onwards: 1.08 - 2019 onwards: 1.02[44] --- Quantitative Factors and Construction 1. Factor Name: Value Factor - **Factor Construction Idea**: Measures stocks with strong trends, low crowding, and moderate odds[27] - **Factor Construction Process**: 1. Trend is measured at zero standard deviation[27] 2. Odds are at 0.3 standard deviation[27] 3. Crowding is at -1.3 standard deviation[27] - **Factor Evaluation**: The factor ranks highest among all style factors, making it a key focus for allocation[27] 2. Factor Name: Quality Factor - **Factor Construction Idea**: Focuses on high odds, weak trends, and low crowding, with potential for future trend confirmation[29] - **Factor Construction Process**: 1. Odds are at 1.7 standard deviation[29] 2. Trend is at -1.4 standard deviation[29] 3. Crowding is at -0.8 standard deviation[29] - **Factor Evaluation**: The factor shows left-side buy signals but requires trend confirmation for stronger allocation[29] 3. Factor Name: Growth Factor - **Factor Construction Idea**: Represents high odds, moderate trends, and moderate crowding, suitable for standard allocation[32] - **Factor Construction Process**: 1. Odds are at 0.9 standard deviation[32] 2. Trend is at -0.2 standard deviation[32] 3. Crowding is at 0.1 standard deviation[32] - **Factor Evaluation**: The factor is recommended for standard allocation due to its balanced characteristics[32] 4. Factor Name: Small-Cap Factor - **Factor Construction Idea**: Characterized by low odds, strong trends, and high crowding, with high uncertainty[35] - **Factor Construction Process**: 1. Odds are at -0.7 standard deviation[35] 2. Trend is at 1.6 standard deviation[35] 3. Crowding is at 0.6 standard deviation[35] - **Factor Evaluation**: The factor is not recommended due to its high uncertainty and crowding[35] --- Factor Backtesting Results 1. Value Factor - **Odds**: 0.3 standard deviation - **Trend**: 0 standard deviation - **Crowding**: -1.3 standard deviation[27] 2. Quality Factor - **Odds**: 1.7 standard deviation - **Trend**: -1.4 standard deviation - **Crowding**: -0.8 standard deviation[29] 3. Growth Factor - **Odds**: 0.9 standard deviation - **Trend**: -0.2 standard deviation - **Crowding**: 0.1 standard deviation[32] 4. Small-Cap Factor - **Odds**: -0.7 standard deviation - **Trend**: 1.6 standard deviation - **Crowding**: 0.6 standard deviation[35]

Quantitative Models and Construction 1. Model Name: Odds Ratio + Win Rate Strategy - **Model Construction Idea**: This strategy combines the odds ratio and win rate metrics to allocate risk budgets across assets, aiming to optimize returns under historical data constraints [3][46] - **Model Construction Process**: - The odds ratio and win rate metrics are calculated for each asset based on historical data - The risk budgets derived from these two metrics are summed to form a composite score - Asset allocation is determined by the composite score, with higher scores receiving higher allocations - Current allocation recommendation: 11.5% equities, 2.2% gold, 86.3% bonds [3][46] - **Model Evaluation**: The model demonstrates stable performance with low drawdowns, making it suitable for risk-averse investors [3][46] 2. Model Name: Odds Ratio Enhanced Strategy - **Model Construction Idea**: Focuses on maximizing returns by overweighting high-odds assets and underweighting low-odds assets under a volatility constraint [40][41] - **Model Construction Process**: - Odds ratios are calculated for each asset - A fixed volatility constraint is applied to ensure risk control - Asset allocation is adjusted dynamically based on odds ratios - Current allocation recommendation: 15.6% equities, 2.9% gold, 81.5% bonds [40][41] - **Model Evaluation**: The strategy effectively balances risk and return, achieving consistent performance over time [40][41] 3. Model Name: Win Rate Enhanced Strategy - **Model Construction Idea**: Utilizes macroeconomic factors (e.g., monetary policy, credit, growth, inflation, and overseas conditions) to derive win rate scores for asset allocation [43][44] - **Model Construction Process**: - Win rate scores are calculated based on macroeconomic indicators - Asset allocation is determined by the win rate scores, favoring assets with higher scores - Current allocation recommendation: 6.6% equities, 1.7% gold, 91.7% bonds [43][44] - **Model Evaluation**: The strategy is robust in capturing macroeconomic trends, providing a defensive allocation approach [43][44] --- Model Backtesting Results 1. Odds Ratio + Win Rate Strategy - Annualized Return: 7.0% (2011–2025), 7.6% (2014–2025), 7.2% (2019–2025) - Maximum Drawdown: 2.8% (2011–2025), 2.7% (2014–2025), 2.8% (2019–2025) - Sharpe Ratio: 2.86 (2011–2025), 3.26 (2014–2025), 2.85 (2019–2025) [3][46][47] 2. Odds Ratio Enhanced Strategy - Annualized Return: 6.6% (2011–2025), 7.5% (2014–2025), 7.0% (2019–2025) - Maximum Drawdown: 3.0% (2011–2025), 2.4% (2014–2025), 2.4% (2019–2025) - Sharpe Ratio: 2.72 (2011–2025), 3.19 (2014–2025), 3.02 (2019–2025) [40][41][42] 3. Win Rate Enhanced Strategy - Annualized Return: 7.0% (2011–2025), 7.7% (2014–2025), 6.3% (2019–2025) - Maximum Drawdown: 2.8% (2011–2025), 2.3% (2014–2025), 2.3% (2019–2025) - Sharpe Ratio: 2.96 (2011–2025), 3.36 (2014–2025), 2.87 (2019–2025) [43][44][45] --- Quantitative Factors and Construction 1. Factor Name: Value Factor - **Factor Construction Idea**: Measures the relative attractiveness of value stocks based on odds, trends, and crowding metrics [18][20] - **Factor Construction Process**: - Odds: 0.2 standard deviations (higher indicates cheaper valuation) - Trend: -0.1 standard deviations (moderate level) - Crowding: -1.0 standard deviations (low crowding) - Composite Score: 1.0 (highest among all factors) [18][20] - **Factor Evaluation**: Strong trend and low crowding make it a top-performing factor [18][20] 2. Factor Name: Quality Factor - **Factor Construction Idea**: Focuses on high-quality stocks with favorable odds and low crowding, awaiting trend confirmation [20][21] - **Factor Construction Process**: - Odds: 1.4 standard deviations (high level) - Trend: -0.3 standard deviations (weak level) - Crowding: -0.8 standard deviations (low level) - Composite Score: 0.6 [20][21] - **Factor Evaluation**: Promising long-term potential but requires trend confirmation for stronger performance [20][21] 3. Factor Name: Growth Factor - **Factor Construction Idea**: Targets growth stocks with improving odds and moderate crowding [23][25] - **Factor Construction Process**: - Odds: 0.6 standard deviations (moderate level) - Trend: 0.02 standard deviations (neutral level) - Crowding: -0.1 standard deviations (moderate level) - Composite Score: 0.4 [23][25] - **Factor Evaluation**: Suitable for neutral allocation due to balanced metrics [23][25] 4. Factor Name: Small-Cap Factor - **Factor Construction Idea**: Captures small-cap stocks with strong trends but high crowding and low odds [26][28] - **Factor Construction Process**: - Odds: -0.5 standard deviations (low level) - Trend: 0.9 standard deviations (high level) - Crowding: 0.6 standard deviations (high level) - Composite Score: 0.0 [26][28] - **Factor Evaluation**: High uncertainty due to low odds and high crowding, requiring cautious approach [26][28] --- Factor Backtesting Results 1. Value Factor - Odds: 0.2 standard deviations - Trend: -0.1 standard deviations - Crowding: -1.0 standard deviations - Composite Score: 1.0 [18][20] 2. Quality Factor - Odds: 1.4 standard deviations - Trend: -0.3 standard deviations - Crowding: -0.8 standard deviations - Composite Score: 0.6 [20][21] 3. Growth Factor - Odds: 0.6 standard deviations - Trend: 0.02 standard deviations - Crowding: -0.1 standard deviations - Composite Score: 0.4 [23][25] 4. Small-Cap Factor - Odds: -0.5 standard deviations - Trend: 0.9 standard deviations - Crowding: 0.6 standard deviations - Composite Score: 0.0 [26][28]

Quantitative Models and Construction 1. Model Name: AIAE Indicator for A-shares - **Model Construction Idea**: The AIAE indicator is used to measure the relative valuation of A-shares by comparing the total market capitalization of the CSI All Share Index with the sum of the total market capitalization and total entity debt[10] - **Model Construction Process**: The formula for the AIAE indicator is: $ AIAE = \frac{\text{CSI All Share Total Market Cap}}{\text{CSI All Share Total Market Cap} + \text{Total Entity Debt}} $ As of the end of May, the AIAE indicator for A-shares was 16%, which is at the 35th percentile since 2010, indicating relatively high valuation attractiveness[10] - **Model Evaluation**: The indicator suggests that A-shares still have high payoff potential, though the win rate remains moderate due to macroeconomic uncertainties[10] 2. Model Name: Bond Payoff Indicator - **Model Construction Idea**: This indicator is derived from the expected return spread between long-term and short-term bonds to assess the valuation risk of bonds[11] - **Model Construction Process**: The bond payoff indicator is calculated based on the expected return difference between long-term and short-term bonds. Currently, the indicator is at -2.1 standard deviations, indicating extremely low valuation levels and potential risks in long-term bonds[11] - **Model Evaluation**: The indicator highlights valuation risks in long-term bonds, though the win rate has improved due to monetary easing and weak credit conditions[11] 3. Model Name: Federal Reserve Liquidity Index - **Model Construction Idea**: This index combines quantity and price dimensions to measure the liquidity provided by the Federal Reserve[18] - **Model Construction Process**: The Federal Reserve Liquidity Index is constructed by integrating multiple factors, including net liquidity, credit support, market expectations, and announcement surprises. Currently, the index is at the 20th percentile, indicating relatively loose liquidity conditions[18] - **Model Evaluation**: The index suggests that liquidity conditions are supportive, but potential shifts in Federal Reserve policy could alter the outlook[18] --- Quantitative Factors and Construction 1. Factor Name: Quality Factor - **Factor Construction Idea**: The quality factor is evaluated based on its payoff, trend, and crowding levels, with a focus on long-term stability[19] - **Factor Construction Process**: - Payoff: Currently at 1.3 standard deviations, indicating attractive valuation - Trend: At -0.3 standard deviations, suggesting moderate momentum - Crowding: At -0.8 standard deviations, reflecting low crowding levels The comprehensive score for the quality factor is 2.4, making it a high-priority allocation[19] - **Factor Evaluation**: The factor is attractive for long-term investment due to its favorable valuation and low crowding[19] 2. Factor Name: Growth Factor - **Factor Construction Idea**: The growth factor is assessed based on its valuation, trend, and crowding, with a focus on growth potential[21] - **Factor Construction Process**: - Payoff: At -1.9 standard deviations, indicating low valuation attractiveness - Trend: At 0.4 standard deviations, suggesting moderate momentum - Crowding: At 0.3 standard deviations, reflecting moderate crowding The comprehensive score for the growth factor is -1.6, indicating low allocation value[21] - **Factor Evaluation**: The factor is less attractive due to its low valuation and moderate crowding[21] 3. Factor Name: Dividend Factor - **Factor Construction Idea**: The dividend factor is evaluated for its income-generating potential and stability[24] - **Factor Construction Process**: - Payoff: At 0.02 standard deviations, indicating neutral valuation - Trend: At -1.8 standard deviations, suggesting weak momentum - Crowding: At -1.2 standard deviations, reflecting low crowding The comprehensive score for the dividend factor is 0, indicating no significant allocation value[24] - **Factor Evaluation**: The factor lacks strong investment appeal due to weak momentum and neutral valuation[24] 4. Factor Name: Small-cap Factor - **Factor Construction Idea**: The small-cap factor is assessed for its potential to outperform based on size and market dynamics[26] - **Factor Construction Process**: - Payoff: At -0.3 standard deviations, indicating neutral valuation - Trend: At 0.4 standard deviations, suggesting moderate momentum - Crowding: At 0.5 standard deviations, reflecting moderate crowding The comprehensive score for the small-cap factor is 0, indicating high uncertainty[26] - **Factor Evaluation**: The factor is not recommended due to its high uncertainty and moderate crowding[26] --- Backtesting Results for Models 1. AIAE Indicator for A-shares - Current value: 16% - Percentile since 2010: 35%[10] 2. Bond Payoff Indicator - Current value: -2.1 standard deviations[11] 3. Federal Reserve Liquidity Index - Current value: 20th percentile[18] --- Backtesting Results for Factors 1. Quality Factor - Payoff: 1.3 standard deviations - Trend: -0.3 standard deviations - Crowding: -0.8 standard deviations - Comprehensive Score: 2.4[19] 2. Growth Factor - Payoff: -1.9 standard deviations - Trend: 0.4 standard deviations - Crowding: 0.3 standard deviations - Comprehensive Score: -1.6[21] 3. Dividend Factor - Payoff: 0.02 standard deviations - Trend: -1.8 standard deviations - Crowding: -1.2 standard deviations - Comprehensive Score: 0[24] 4. Small-cap Factor - Payoff: -0.3 standard deviations - Trend: 0.4 standard deviations - Crowding: 0.5 standard deviations - Comprehensive Score: 0[26]