Quantitative Models and Construction A-Share Market Timing Model - **Model Name**: A-Share Multi-Dimensional Timing Model [10] - **Construction Idea**: The model integrates valuation, sentiment, capital, and technical dimensions to assess the directional outlook of the A-share market [10][12][16] - **Construction Process**: - Signals are generated daily for each dimension, with values of 0, ±1 representing neutral, bullish, and bearish views respectively [10] - **Valuation Dimension**: Uses equity risk premium (ERP) to capture mean-reversion characteristics [12][16] - **Sentiment Dimension**: Includes option put-call ratio, implied volatility, and futures member position ratio to reflect market sentiment [12][16] - **Capital Dimension**: Tracks financing purchase amounts to identify market trends [12][16] - **Technical Dimension**: Employs Bollinger Bands and individual stock turnover ratio differences to capture trend continuation [12][16] - The final market view is determined by the sum of scores across all dimensions [10] - **Evaluation**: The model effectively combines mean-reversion and trend-following strategies, balancing risk avoidance and opportunity capture [10] Style Timing Model - **Model Name**: Dividend Style Timing Model [18] - **Construction Idea**: Targets the relative performance of the CSI Dividend Index against the CSI All Index using trend-based indicators [18][22] - **Construction Process**: - Three indicators are used to generate daily signals (0, ±1 for neutral, bullish, bearish views) [18] - **Relative Momentum**: Positive indicator for dividend style [22] - **10Y-1Y Term Spread**: Negative indicator for dividend style, as wider spreads favor growth assets [22] - **Interbank Repo Volume**: Positive indicator for dividend style, reflecting asset scarcity [22] - Signals are aggregated to determine the overall view on dividend style [18] - **Evaluation**: The model captures dividend style trends effectively, leveraging macroeconomic and liquidity factors [18] - **Model Name**: Large-Cap vs Small-Cap Style Timing Model [23] - **Construction Idea**: Differentiates between macro-driven trends in low congestion and fund-driven reversals in high congestion [23][25] - **Construction Process**: - **Momentum Difference**: Calculates the difference in momentum between the Wind Micro-Cap Index and CSI 300 Index across multiple windows, averaging the top/bottom results for small/large-cap scores [27] - **Turnover Ratio**: Similar calculation for turnover ratio differences across windows, averaged for small/large-cap scores [27] - **Congestion Score**: Combines momentum and turnover scores to determine congestion levels (high congestion >90% for small-cap, <10% for large-cap) [27] - **Trend Model**: Uses small/large parameter double moving average models based on congestion levels [25] - **Evaluation**: The model adapts to market conditions, balancing long-term trends and short-term reversals [23][25] Sector Rotation Model - **Model Name**: Genetic Programming Sector Rotation Model [30] - **Construction Idea**: Directly mines factors from sector index data using genetic programming without relying on predefined scoring rules [30][33] - **Construction Process**: - **Factor Mining**: Utilizes NSGA-II algorithm to optimize for monotonicity and top-group performance simultaneously [33][34] - **Factor Combination**: Combines factors with weak collinearity using greedy strategy and variance inflation coefficient [34] - **Weekly Rebalancing**: Selects top five sectors based on multi-factor scores for equal-weight allocation [30] - **Example Factor**: Calculates covariance between standardized weekly low prices and monthly open prices over 25 days, adjusted by standardized weekly high prices over 15 days [38] - **Evaluation**: The model enhances factor diversity and reduces overfitting risks, achieving robust sector rotation performance [33][34] All-Weather Enhanced Portfolio - **Model Name**: China All-Weather Enhanced Portfolio [39] - **Construction Idea**: Implements macro factor risk parity to diversify risks across underlying macro drivers rather than assets [39][42] - **Construction Process**: - **Macro Quadrant Division**: Divides growth and inflation dimensions into four quadrants based on whether they exceed or fall short of expectations [42] - **Quadrant Portfolio Construction**: Constructs sub-portfolios within each quadrant, focusing on downside risk [42] - **Risk Budgeting**: Adjusts quadrant weights monthly based on macro momentum indicators combining buy-side and sell-side expectations [42] - **Evaluation**: The strategy demonstrates strong defensive attributes during market downturns while maintaining consistent returns [40][43] --- Backtesting Results A-Share Market Timing Model - **Annualized Return**: 24.94% [15] - **Maximum Drawdown**: -28.46% [15] - **Sharpe Ratio**: 1.16 [15] - **Calmar Ratio**: 0.88 [15] - **YTD Return**: 43.84% [15] - **Weekly Return**: 5.28% [15] Dividend Style Timing Model - **Annualized Return**: 15.67% [21] - **Maximum Drawdown**: -25.52% [21] - **Sharpe Ratio**: -0.26 [21] - **Calmar Ratio**: 0.85 [21] - **YTD Return**: 20.86% [21] - **Weekly Return**: -3.63% [21] Large-Cap vs Small-Cap Style Timing Model - **Annualized Return**: 27.04% [28] - **Maximum Drawdown**: -32.05% [28] - **Sharpe Ratio**: 1.13 [28] - **Calmar Ratio**: 0.84 [28] - **YTD Return**: 71.14% [28] - **Weekly Return**: -7.80% [28] Sector Rotation Model - **Annualized Return**: 30.83% [33] - **Annualized Volatility**: 17.74% [33] - **Sharpe Ratio**: 1.74 [33] - **Maximum Drawdown**: -19.63% [33] - **Calmar Ratio**: 1.57 [33] - **YTD Return**: 35.44% [33] - **Weekly Return**: -4.39% [33] All-Weather Enhanced Portfolio - **Annualized Return**: 11.51% [43] - **Annualized Volatility**: 6.18% [43] - **Sharpe Ratio**: 1.86 [43] - **Maximum Drawdown**: -6.30% [43] - **Calmar Ratio**: 1.83 [43] - **YTD Return**: 10.75% [43] - **Weekly Return**: -1.53% [43]

- The "A-Share Market Timing Model" evaluates the overall directional judgment of the A-share market using four dimensions: valuation, sentiment, funds, and technicals. The model generates daily signals with values of 0, ±1, representing neutral, bullish, and bearish views, respectively. The model's logic includes mean reversion for valuation and sentiment, and trend continuation for funds and technicals[2][9][15] - The "Style Timing Model" favors a barbell structure of dividend and small-cap styles. For the dividend style, the model uses the relative momentum of the CSI Dividend Index to the CSI All Share Index, the 10Y-1Y term spread, and the interbank pledged repo transaction volume. For the small-cap style, the model employs a trend model based on the difference in momentum and trading volume between small-cap and large-cap stocks[3][17][21] - The "Industry Rotation Model" uses genetic programming to directly extract factors from the volume, price, and valuation characteristics of industry indices. The model updates its factor library quarterly and rebalances weekly, selecting the top five industries with the highest multi-factor composite scores for equal-weight allocation[4][29][34] - The "China Domestic All-Weather Enhanced Portfolio" employs a macro factor risk parity framework, selecting four macro risk sources: growth above/below expectations and inflation above/below expectations. The model actively overweights favored quadrants based on macro expectation momentum, adjusting monthly[5][39][42] - The "A-Share Market Timing Model" achieved a year-to-date return of 36.03%, with an excess return of 8.86% over the Wind All A Index, which had a return of 27.18%[2][9] - The "Style Timing Model" for the dividend style yielded a year-to-date return of 25.04%, with an excess return of 7.83% over the benchmark, which had a return of 17.21%[17][20] - The "Style Timing Model" for the small-cap style achieved a year-to-date return of 78.29%, with an excess return of 30.25% over the benchmark, which had a return of 48.04%[22][27] - The "Industry Rotation Model" achieved a year-to-date return of 40.67%, outperforming the industry equal-weight benchmark by 17.96 percentage points[4][32] - The "China Domestic All-Weather Enhanced Portfolio" achieved a year-to-date return of 11.10%, with a Sharpe ratio of 2.22, a maximum drawdown of 2.67%, and a Calmar ratio of 5.15[5][40][43]

- **A-share multi-dimensional timing model**: The model evaluates the overall directional judgment of the A-share market using four dimensions: valuation, sentiment, funds, and technical indicators. Each dimension provides daily signals with values of 0, ±1, representing neutral, bullish, or bearish views. Valuation and sentiment dimensions adopt a mean-reversion logic, while funds and technical dimensions use trend-following logic. The final market view is determined by the sum of the scores across all dimensions [9][15][16] - **Style timing model for dividend style**: The model uses three indicators to time the dividend style relative to the CSI Dividend Index and CSI All Share Index. The indicators include relative momentum, 10Y-1Y term spread, and interbank pledged repo transaction volume. Each indicator provides daily signals with values of 0, ±1, representing neutral, bullish, or bearish views. The final view is based on the sum of the scores across all dimensions. When the model favors the dividend style, it fully allocates to the CSI Dividend Index; otherwise, it allocates to the CSI All Share Index [17][21] - **Style timing model for large-cap and small-cap styles**: The model uses momentum difference and turnover ratio difference between the CSI 300 Index and Wind Micro Cap Index to calculate the crowding scores for large-cap and small-cap styles. The model operates in two crowding zones: high crowding and low crowding. In high crowding zones, it uses a small-parameter dual moving average model to address potential style reversals. In low crowding zones, it uses a large-parameter dual moving average model to capture medium- to long-term trends [22][24][26] - **Sector rotation model**: The genetic programming-based sector rotation model selects the top five sectors with the highest multi-factor composite scores from 32 CITIC industry indices for equal-weight allocation. The model updates its factor library quarterly and rebalances weekly. The factors are derived using NSGA-II algorithm, which evaluates factor monotonicity and performance of long positions using |IC| and NDCG@5 metrics. The model combines multiple factors with weak collinearity into sector scores using greedy strategy and variance inflation factor [29][32][33][36] - **China domestic all-weather enhanced portfolio**: The portfolio is constructed using a macro factor risk parity framework, which emphasizes risk diversification across underlying macro risk sources rather than asset classes. The strategy involves three steps: macro quadrant classification and asset selection, quadrant portfolio construction and risk measurement, and risk budgeting to determine quadrant weights. The active allocation is based on macro expectation momentum indicators, which consider buy-side expectation momentum and sell-side expectation deviation momentum [38][41] --- Model Backtesting Results - **A-share multi-dimensional timing model**: Annualized return 24.97%, maximum drawdown -28.46%, Sharpe ratio 1.16, Calmar ratio 0.88, YTD return 37.73%, weekly return 0.00% [14] - **Dividend style timing model**: Annualized return 15.71%, maximum drawdown -25.52%, Sharpe ratio 0.85, Calmar ratio 0.62, YTD return 19.53%, weekly return -3.43% [20] - **Large-cap vs. small-cap style timing model**: Annualized return 26.01%, maximum drawdown -30.86%, Sharpe ratio 1.08, Calmar ratio 0.84, YTD return 64.58%, weekly return -2.22% [27] - **Sector rotation model**: Annualized return 33.33%, annualized volatility 17.89%, Sharpe ratio 1.86, maximum drawdown -19.63%, Calmar ratio 1.70, weekly return 0.14%, YTD return 39.41% [32] - **China domestic all-weather enhanced portfolio**: Annualized return 11.66%, annualized volatility 6.18%, Sharpe ratio 1.89, maximum drawdown -6.30%, Calmar ratio 1.85, weekly return 0.38%, YTD return 10.74% [42]