- The report tracks high-frequency stock selection factors, including Price Range Factor, Price-Volume Divergence Factor, Regret Avoidance Factor, and Slope Convexity Factor, with their out-of-sample performance showing overall excellence[2][3][11] - Price Range Factor measures the activity level of stocks traded within different intraday price ranges, reflecting investors' expectations for future stock trends. It demonstrates strong predictive power and stable performance this year[3][11][17] - Price-Volume Divergence Factor evaluates the correlation between stock prices and trading volumes. Lower correlation typically indicates higher potential for future stock price increases. However, its performance has been unstable in recent years, with multi-long net value curves flattening[3][22][26] - Regret Avoidance Factor examines the proportion and degree of stock rebounds after being sold by investors, showcasing good predictive power. Its out-of-sample excess returns are stable, indicating that A-share investors' regret avoidance sentiment significantly impacts stock price expectations[3][27][36] - Slope Convexity Factor analyzes the slope and convexity of order books to assess the impact of investor patience and supply-demand elasticity on expected returns. It is constructed using high-frequency snapshot data from limit order books[3][37][42] - The report combines three high-frequency factors into an equal-weighted "Gold" portfolio for CSI 1000 Index enhancement strategy, achieving an annualized excess return rate of 10.51% and a maximum excess drawdown of 6.04%[3][44][45] - To further enhance strategy performance, the report integrates high-frequency factors with three effective fundamental factors (Consensus Expectations, Growth, and Technical Factors) to construct a high-frequency & fundamental resonance portfolio for CSI 1000 Index enhancement strategy. This strategy achieves an annualized excess return rate of 14.57% and a maximum excess drawdown of 4.52%[4][49][51] Factor Backtesting Results - Price Range Factor: Weekly excess return 0.40%, monthly excess return 0.51%, annual excess return 5.86%[2][13][17] - Price-Volume Divergence Factor: Weekly excess return -0.24%, monthly excess return 1.53%, annual excess return 9.00%[2][13][26] - Regret Avoidance Factor: Weekly excess return 0.27%, monthly excess return -0.49%, annual excess return 2.32%[2][13][36] - Slope Convexity Factor: Weekly excess return -1.74%, monthly excess return -2.46%, annual excess return -5.90%[2][13][42] Strategy Performance Metrics - "Gold" Portfolio: Annualized return 9.49%, annualized excess return 10.51%, Sharpe ratio 0.39, IR 2.47, maximum excess drawdown 6.04%[45][47][48] - High-frequency & Fundamental Resonance Portfolio: Annualized return 13.62%, annualized excess return 14.57%, Sharpe ratio 0.58, IR 3.50, maximum excess drawdown 4.52%[51][53][55]

Core Insights - The report highlights the significant performance of the A-share market driven by passive investment and leveraged funds, with the total margin financing and securities lending balance exceeding 1.99 trillion as of August 4, 2025, marking a historical high since 2024 [5][8][6] - The automotive sector, particularly the company North Car Blue Valley (600733.SH), has launched a "Three-Year Leap Plan" aimed at enhancing profitability through sales growth, structural optimization, cost control, and expanding its profit ecosystem [4][16] - The company reported a 151% year-on-year increase in revenue for Q1 2025, with a gross margin improvement of 4.1 percentage points, and a reduction in net loss by 60 million [4][16] Industry Overview - The automotive industry is focusing on high-end market penetration, with North Car Blue Valley collaborating with Huawei to enhance its brand image and product offerings, particularly in the high-end vehicle segment [18][17] - The report indicates a notable increase in sales for the "Extreme Fox" brand due to comprehensive adjustments in product positioning, marketing strategies, and channel expansion [17] - The "Enjoy" brand, under the Huawei partnership, aims to redefine high-end sedans with innovative features and improved range, which is expected to boost sales significantly [18] Market Dynamics - The report discusses the microstructure of the market, emphasizing the importance of early trading concentration and the dynamics between institutional and retail investors [9][10][12] - It notes that the market's profitability effect has increased retail participation, contrasting with the trend of rising institutional ownership since 2017 [6][8] - The report tracks high-frequency factors, indicating strong performance in various trading strategies, with notable returns from specific factors such as the high-dimensional memory factor yielding 29.3% since 2023 [14]

Quantitative Models and Construction Methods - **Model Name**: High-dimensional Memory (MEMO) Factor **Construction Idea**: This factor uses symbol processing to analyze the relationship between each order and subsequent orders, reflecting institutional contributions to trading[40][45] **Construction Process**: 1. Convert the trading direction of each order into a numerical sequence 2. Calculate the correlation coefficient between orders to measure their relationship 3. Stronger correlations indicate higher institutional involvement and better company quality[40][45] **Evaluation**: The factor effectively captures institutional trading behavior and demonstrates strong performance in identifying high-quality stocks[40][45] - **Model Name**: Strong Reversal (SR) Factor **Construction Idea**: Based on the principle that higher single-order transaction amounts lead to stronger reversals, this factor refines the ideal reversal factor at the minute level[46][48] **Construction Process**: 1. Use minute-level single-order transaction amounts 2. Segment the intraday 240-minute price fluctuations 3. Construct the strong reversal factor based on the ideal reversal factor[46][48] **Evaluation**: The factor improves upon daily frequency reversal factors and effectively captures intraday reversal opportunities[46][48] - **Model Name**: Lottery (LOTTERY) Factor **Construction Idea**: This factor identifies retail investor behavior by analyzing orders placed at limit-up or limit-down prices, reflecting the dominance of retail characteristics in trading[48][49] **Construction Process**: 1. Analyze the proportion of orders placed at limit-up or limit-down prices 2. Higher proportions indicate retail-dominated trading structures 3. Stocks with higher retail dominance often exhibit price deviations[48][49] **Evaluation**: The factor effectively captures retail investor behavior and highlights stocks with potential price anomalies[48][49] Model Backtesting Results - **MEMO Factor**: - IC: 0.045 - ICIR: 2.989 - Annualized Long-Short Return: 29.3%[39][40][45] - **SR Factor**: - IC: -0.043 - ICIR: -2.473 - Annualized Long-Short Return: 19.7%[39][46][48] - **LOTTERY Factor**: - IC: -0.054 - ICIR: -2.792 - Annualized Long-Short Return: 32.9%[39][48][49] High-Frequency Factor Tracking Results - **MEMO Factor**: - IC: 0.045 - ICIR: 2.989 - Annualized Long-Short Return: 29.3%[39][40][45] - **SR Factor**: - IC: -0.043 - ICIR: -2.473 - Annualized Long-Short Return: 19.7%[39][46][48] - **LOTTERY Factor**: - IC: -0.054 - ICIR: -2.792 - Annualized Long-Short Return: 32.9%[39][48][49]

Quantitative Models and Construction Methods Quantitative Factors and Their Construction 1. **Factor Name**: Intraday Skewness Factor **Construction Idea**: Captures the skewness of intraday stock returns to identify potential return asymmetry[3][6] **Construction Process**: Referenced in the report "Stock Selection Factor Series Research (19) - High-Frequency Factors on Stock Return Distribution Characteristics"[11] **Evaluation**: Demonstrates mixed performance with positive returns in some periods but underperformance in others[3][6] 2. **Factor Name**: Downside Volatility Proportion Factor **Construction Idea**: Measures the proportion of downside volatility in intraday price movements to assess risk[3][6] **Construction Process**: Referenced in the report "Stock Selection Factor Series Research (25) - High-Frequency Factors on Realized Volatility Decomposition"[16] **Evaluation**: Shows consistent positive returns in certain periods but limited robustness in others[3][6] 3. **Factor Name**: Post-Open Buy Intention Proportion Factor **Construction Idea**: Quantifies the proportion of buy orders after market open to gauge investor sentiment[3][6] **Construction Process**: Referenced in the report "Stock Selection Factor Series Research (64) - Low-Frequency Applications of High-Frequency Data Using Intuitive Logic and Machine Learning"[20] **Evaluation**: Exhibits moderate performance with occasional strong returns[3][6] 4. **Factor Name**: Post-Open Buy Intention Intensity Factor **Construction Idea**: Measures the intensity of buy orders after market open to reflect market momentum[3][6] **Construction Process**: Referenced in the report "Stock Selection Factor Series Research (64) - Low-Frequency Applications of High-Frequency Data Using Intuitive Logic and Machine Learning"[24] **Evaluation**: Performance is inconsistent, with periods of underperformance[3][6] 5. **Factor Name**: Post-Open Large Order Net Buy Proportion Factor **Construction Idea**: Tracks the proportion of large net buy orders after market open to identify institutional activity[3][6] **Construction Process**: Derived from high-frequency trading data[30] **Evaluation**: Generally positive performance with strong returns in specific periods[3][6] 6. **Factor Name**: Post-Open Large Order Net Buy Intensity Factor **Construction Idea**: Measures the intensity of large net buy orders after market open to capture market trends[3][6] **Construction Process**: Derived from high-frequency trading data[35] **Evaluation**: Mixed results with moderate returns in some periods[3][6] 7. **Factor Name**: Improved Reversal Factor **Construction Idea**: Enhances traditional reversal factors by incorporating high-frequency data[3][6] **Construction Process**: Derived from intraday price reversals[40] **Evaluation**: Limited performance improvement over traditional reversal factors[3][6] 8. **Factor Name**: Tail-End Trading Proportion Factor **Construction Idea**: Measures the proportion of trading activity near market close to capture end-of-day effects[3][6] **Construction Process**: Derived from high-frequency trading data[45] **Evaluation**: Underperformance in most periods[3][6] 9. **Factor Name**: Average Single Transaction Outflow Proportion Factor **Construction Idea**: Tracks the proportion of outflows in single transactions to assess liquidity[3][6] **Construction Process**: Derived from high-frequency trading data[50] **Evaluation**: Limited effectiveness in predicting returns[3][6] 10. **Factor Name**: Large Order Push-Up Factor **Construction Idea**: Measures the impact of large orders on price increases to identify market movers[3][6] **Construction Process**: Derived from high-frequency trading data[55] **Evaluation**: Moderate performance with occasional strong returns[3][6] 11. **Factor Name**: Deep Learning High-Frequency Factor (Improved GRU(50,2)+NN(10)) **Construction Idea**: Combines GRU and neural networks to capture complex patterns in high-frequency data[3][6] **Construction Process**: Utilizes GRU(50,2) and NN(10) architectures for feature extraction and prediction[59] **Evaluation**: Strong performance in certain periods but underperformance in others[3][6] 12. **Factor Name**: Deep Learning High-Frequency Factor (Residual Attention LSTM(48,2)+NN(10)) **Construction Idea**: Incorporates residual attention mechanisms with LSTM and neural networks for enhanced prediction[3][6] **Construction Process**: Utilizes LSTM(48,2) and NN(10) architectures with residual attention layers[61] **Evaluation**: Consistently strong performance across multiple periods[3][6] 13. **Factor Name**: Deep Learning Factor (Multi-Granularity Model - 5-Day Label) **Construction Idea**: Uses multi-granularity modeling with 5-day labels for short-term predictions[3][6] **Construction Process**: Trained using bidirectional AGRU[64] **Evaluation**: Strong performance with high returns in most periods[3][6] 14. **Factor Name**: Deep Learning Factor (Multi-Granularity Model - 10-Day Label) **Construction Idea**: Uses multi-granularity modeling with 10-day labels for medium-term predictions[3][6] **Construction Process**: Trained using bidirectional AGRU[65] **Evaluation**: Consistently strong performance across multiple periods[3][6] AI-Enhanced Portfolio Construction 1. **Portfolio Name**: CSI 500 AI Enhanced Wide Constraint Portfolio **Construction Idea**: Maximizes expected returns under wide constraints using deep learning factors[69][70] **Construction Process**: - Weekly rebalancing - Constraints on individual stocks, industries, market cap, and other factors - Objective function: $$ max\sum\mu_{i}w_{i} $$ where \( w_i \) is the weight of stock \( i \) and \( \mu_i \) is its expected excess return[71] **Evaluation**: Strong cumulative excess returns since 2017[72] 2. **Portfolio Name**: CSI 500 AI Enhanced Strict Constraint Portfolio **Construction Idea**: Similar to the wide constraint portfolio but with stricter constraints[69][70] **Construction Process**: Same as above with stricter constraints on market cap, ROE, SUE, and volatility[71] **Evaluation**: Moderate cumulative excess returns since 2017[73] 3. **Portfolio Name**: CSI 1000 AI Enhanced Wide Constraint Portfolio **Construction Idea**: Maximizes expected returns under wide constraints using deep learning factors for smaller-cap stocks[69][70] **Construction Process**: Same as CSI 500 portfolios but applied to CSI 1000 index[71] **Evaluation**: Strong cumulative excess returns since 2017[76] 4. **Portfolio Name**: CSI 1000 AI Enhanced Strict Constraint Portfolio **Construction Idea**: Similar to the wide constraint portfolio but with stricter constraints for smaller-cap stocks[69][70] **Construction Process**: Same as above with stricter constraints on market cap, ROE, SUE, and volatility[71] **Evaluation**: Strong cumulative excess returns since 2017[79] Backtest Results for Factors 1. **Intraday Skewness Factor**: IC (2025): 0.057, Multi-Period Returns: 14.35% (2025)[3][6] 2. **Downside Volatility Proportion Factor**: IC (2025): 0.055, Multi-Period Returns: 11.77% (2025)[3][6] 3. **Post-Open Buy Intention Proportion Factor**: IC (2025): 0.033, Multi-Period Returns: 10.32% (2025)[3][6] 4. **Post-Open Buy Intention Intensity Factor**: IC (2025): 0.026, Multi-Period Returns: 11.19% (2025)[3][6] 5. **Post-Open Large Order Net Buy Proportion Factor**: IC (2025): 0.039, Multi-Period Returns: 12.32% (2025)[3][6] 6. **Post-Open Large Order Net Buy Intensity Factor**: IC (2025): 0.028, Multi-Period Returns: 6.78% (2025)[3][6] 7. **Improved Reversal Factor**: IC (2025): 0.003, Multi-Period Returns: 9.34% (2025)[3][6] 8. **Tail-End Trading Proportion Factor**: IC (2025): 0.022, Multi-Period Returns: 5.43% (2025)[3][6] 9. **Average Single Transaction Outflow Proportion Factor**: IC (2025): 0.012, Multi-Period Returns: 0.82% (2025)[3][6] 10. **Large Order Push-Up Factor

Group 1: ETF Rotation Strategy Performance - The ETF rotation strategy, constructed using GBDT+NN machine learning factors, has shown excellent out-of-sample performance with an IC value of 44.48% and a long position excess return of 0.73% last week [3][14] - The annualized excess return of the strategy is 11.88%, with a maximum drawdown of 17.31% [17][18] - Recent performance includes an excess return of 0.20% last week, 1.64% for the month, and 0.35% year-to-date [18][20] Group 2: High-Frequency Factor Overview - Various high-frequency factors have demonstrated strong overall performance, with the price range factor showing a long position excess return of 4.93% year-to-date, while the regret avoidance factor has underperformed with a return of 0.27% [4][22] - The price range factor measures the activity level of stocks within different price ranges, indicating investor expectations for future price movements [5][25] - The regret avoidance factor reflects the impact of investor emotions on stock price expectations, showing stable out-of-sample excess returns [5][37] Group 3: High-Frequency and Fundamental Factor Combination - A combined strategy of high-frequency and fundamental factors has been developed, yielding an annualized excess return of 14.76% with a maximum drawdown of 4.52% [6][59] - The strategy has shown stable out-of-sample performance, with a year-to-date excess return of 3.74% [60] - The integration of fundamental factors with high-frequency factors has improved the performance metrics of the strategy [57][59]

Group 1: ETF Rotation Strategy Tracking - The ETF rotation strategy, constructed using GBDT+NN machine learning factors, has shown strong performance in out-of-sample testing, with an annualized excess return of 11.90% and a maximum drawdown of 17.31% [2][12][17] - Recent performance indicates a weekly excess return of 0.77% and a monthly excess return of 1.10%, while the year-to-date excess return stands at -0.19% [20][24] - The strategy's information ratio is 0.68, reflecting its effectiveness in generating excess returns relative to risk [24] Group 2: High-Frequency Factor Overview - High-frequency factors have demonstrated overall strong performance, with the price range factor yielding a year-to-date excess return of 4.79% and the price-volume divergence factor achieving 10.08% [3][20] - The regret avoidance factor has underperformed with a year-to-date excess return of -0.56%, while the slope convexity factor has shown a year-to-date excess return of -3.64% [3][20] - The high-frequency "gold" combination strategy has an annualized excess return of 10.69% and a maximum drawdown of 6.04% [5][60] Group 3: High-Frequency Factor Performance Tracking - The price range factor measures the activity level of stocks within different price ranges, showing strong predictive power and stable performance this year [4][28] - The price-volume divergence factor assesses the correlation between stock price and trading volume, with recent performance indicating a mixed stability [4][39] - The regret avoidance factor reflects investor behavior, showing stable out-of-sample excess returns, while the slope convexity factor illustrates the impact of order book elasticity on expected returns [4][51] Group 4: Combined Strategies Performance - The high-frequency and fundamental resonance combination strategy has an annualized excess return of 14.98% and a maximum drawdown of 4.52% [5][64] - Recent performance for this combined strategy includes a weekly excess return of 0.63% and a monthly excess return of 2.00%, with a year-to-date excess return of 4.69% [67]