Quantitative Models and Construction Methods 1. Model Name: Industry Crowding Monitoring Model - **Model Construction Idea**: This model is designed to monitor the crowding levels of Shenwan First-Level Industry Indices on a daily basis, identifying industries with high or low crowding levels[4] - **Model Construction Process**: The model calculates the crowding levels of various industries based on specific metrics (not detailed in the report) and ranks them accordingly. For the previous trading day, industries such as steel, building materials, and non-ferrous metals had high crowding levels, while media, home appliances, and automobiles had lower levels[4] - **Model Evaluation**: The model provides a useful tool for identifying industry crowding trends and potential investment opportunities or risks[4] 2. Model Name: Premium Rate Z-Score Model - **Model Construction Idea**: This model identifies potential arbitrage opportunities in ETF products by calculating the Z-score of premium rates over a rolling window[5] - **Model Construction Process**: The Z-score is calculated as follows: $ Z = \frac{(P - \mu)}{\sigma} $ where: - $ P $ represents the premium rate of the ETF - $ \mu $ is the mean premium rate over the rolling window - $ \sigma $ is the standard deviation of the premium rate over the rolling window The model flags ETFs with significant deviations from their historical premium rates, indicating potential arbitrage opportunities[5] - **Model Evaluation**: The model is effective in identifying ETFs with potential mispricing but requires caution due to the risk of price corrections[5] --- Backtesting Results of Models 1. Industry Crowding Monitoring Model - No specific numerical backtesting results were provided for this model[4] 2. Premium Rate Z-Score Model - No specific numerical backtesting results were provided for this model[5] --- Quantitative Factors and Construction Methods No specific quantitative factors were detailed in the report. --- Backtesting Results of Factors No specific quantitative factor backtesting results were provided in the report.

Core Viewpoint - The market is expected to experience a correction in the upcoming week due to various technical and quantitative indicators suggesting a weakening market sentiment [1][2]. Market Analysis - The liquidity shock indicator for the CSI 300 index was recorded at 1.71, indicating that current market liquidity is 1.71 times higher than the average level over the past year [2]. - The PUT-CALL ratio for the SSE 50 ETF options increased to 0.80, reflecting a growing caution among investors regarding the short-term performance of the SSE 50 ETF [2]. - The five-day average turnover rates for the SSE Composite Index and Wind All A Index were 1.07% and 1.65%, respectively, indicating a decrease in trading activity [2]. Macroeconomic Factors - The onshore and offshore RMB exchange rates experienced slight declines of -0.08% and -0.1% respectively [2]. - New RMB loans in June amounted to 22,400 billion, exceeding the consensus forecast of 18,447.29 billion and the previous value of 6,200 billion [2]. - The broad money supply (M2) grew by 8.3% year-on-year, surpassing both the consensus forecast of 8.08% and the previous value of 7.9% [2]. Technical Analysis - The Wind All A Index remains above the SAR point, but the index and SAR point are now closely aligned [2]. - The moving average strength index is currently at 253, placing it in the 93.8 percentile since 2021 [2]. - The sentiment model score is 1 out of 5, indicating a decrease in market sentiment, while the trend model signal is positive and the weighted model signal is negative [2]. Performance Overview - For the week of July 14-18, the SSE 50 Index rose by 0.28%, the CSI 300 Index increased by 1.09%, the CSI 500 Index gained 1.2%, and the ChiNext Index surged by 3.17% [3]. - The overall market PE (TTM) stands at 20.4 times, which is at the 65.3 percentile since 2005 [3]. Factor Crowding Observation - The small-cap factor crowding is at a high level with a score of 1.07, while the low valuation factor crowding is at 0.36 [3]. - The industry crowding levels are relatively high in banking, comprehensive, non-ferrous metals, steel, and non-bank financial sectors, with notable increases in steel and pharmaceutical industries [3].

- The report uses the Mahalanobis distance of weekly price and trading volume changes to measure industry crowding levels[3][17] - The construction process involves identifying industries in the first quadrant (price and volume both rising) and the third quadrant (price and volume both falling) and marking points outside the ellipse as industries with significant short-term deviations at a 99% confidence level[17] - The building materials industry experienced short-term trading overselling last week[3][18]

Quantitative Models and Construction Methods 1. Model Name: Industry Crowding Monitoring Model - **Model Construction Idea**: This model is designed to monitor the crowding levels of Shenwan First-Level Industry Indices on a daily basis, identifying industries with high or low crowding levels to provide actionable insights[4] - **Model Construction Process**: The model calculates crowding levels for each industry index daily, using metrics such as main fund flows and single-day crowding changes. For example, the model identified that non-ferrous metals and steel had high crowding levels, while automobiles and electronics had lower levels. Additionally, significant single-day crowding changes were observed in the power equipment sector[4] - **Model Evaluation**: The model provides a useful tool for identifying industry crowding trends and potential investment opportunities[4] 2. Model Name: Premium Rate Z-Score Model - **Model Construction Idea**: This model is used to screen ETF products for potential arbitrage opportunities by calculating the Z-score of premium rates on a rolling basis[5] - **Model Construction Process**: The Z-score is calculated for the premium rates of ETF products over a rolling window. This helps identify ETFs with significant deviations from their historical averages, signaling potential arbitrage opportunities. The model also flags ETFs with potential downside risks[5] - **Model Evaluation**: The model effectively identifies ETFs with potential arbitrage opportunities while also highlighting associated risks[5] --- Model Backtesting Results 1. Industry Crowding Monitoring Model - **Key Observations**: - Non-ferrous metals and steel had the highest crowding levels on the previous trading day[4] - Automobiles and electronics exhibited the lowest crowding levels[4] - Power equipment showed significant single-day crowding changes[4] 2. Premium Rate Z-Score Model - **Key Observations**: - The model identified ETFs with significant premium rate deviations, signaling potential arbitrage opportunities[5] --- Quantitative Factors and Construction Methods No specific quantitative factors were explicitly mentioned in the provided content --- Factor Backtesting Results No specific factor backtesting results were explicitly mentioned in the provided content

- The report constructs an industry crowding model to monitor the daily crowding levels of Shenwan primary industry indices, identifying high crowding in building materials and electrical equipment, while home appliances and transportation show lower levels[3] - A Z-score model is used to screen ETF products based on premium rates, providing signals for potential arbitrage opportunities and warning of potential risks of price corrections[4] - The industry crowding model highlights significant daily changes in crowding levels for agriculture, forestry, animal husbandry, and fishery, as well as non-ferrous metals[3] - The Z-score model applies rolling calculations to identify ETFs with potential arbitrage opportunities, focusing on premium rate deviations[4] - The industry crowding model suggests monitoring industries with extreme crowding levels for potential investment opportunities or risks[3] - The Z-score model emphasizes the importance of tracking premium rate deviations to identify arbitrage opportunities and mitigate risks[4]

- The industry crowding monitoring model was constructed to monitor the daily crowding levels of Shenwan primary industry indices. It identified utilities and building materials as having high crowding levels, while automotive, food & beverage, and home appliances showed low crowding levels. The model also tracked significant daily changes in crowding levels for industries like agriculture, coal, and environmental protection[4] - The Z-score premium rate model was developed to screen ETF products for potential arbitrage opportunities. This model uses rolling calculations to identify signals and warns of potential risks of price corrections for the identified ETFs[5] - Daily net inflows for broad-based ETFs amounted to 3.77 billion yuan, with top inflows observed in CSI 1000 ETF (+7.78 billion yuan), SSE 50 ETF (+6.96 billion yuan), and CSI 300 ETF (+5.38 billion yuan). Conversely, top outflows were recorded for ChiNext ETF (-6.73 billion yuan), CSI A500 ETF (-4.06 billion yuan), and STAR 50 ETF (-3.51 billion yuan)[6] - Industry-themed ETFs saw a daily net inflow of 1.82 billion yuan, with top inflows in Military ETF (+4.01 billion yuan), Securities ETF (+2.63 billion yuan), and Defense ETF (+2.31 billion yuan). Top outflows were noted for Robotics ETF (-1.39 billion yuan), Semiconductor ETF (-1.05 billion yuan), and AI ETF (-0.99 billion yuan)[6] - Style-strategy ETFs recorded a daily net inflow of 2.29 billion yuan, with top inflows in Low Volatility Dividend ETF (+1.62 billion yuan), Low Volatility Dividend 50 ETF (+0.53 billion yuan), and Dividend State-Owned Enterprise ETF (+0.28 billion yuan). Top outflows included CSI Dividend ETF (-0.19 billion yuan), Low Volatility Dividend ETF (-0.18 billion yuan), and Low Volatility Dividend 100 ETF (-0.15 billion yuan)[6] - Cross-border ETFs experienced a daily net outflow of 0.51 billion yuan, with top inflows in Hong Kong Non-Bank ETF (+3.84 billion yuan), Hang Seng Low Volatility Dividend ETF (+0.63 billion yuan), and S&P 500 ETF (+0.42 billion yuan). Top outflows were observed for Hang Seng Tech ETF (-1.19 billion yuan), Hong Kong Dividend ETF (-0.82 billion yuan), and Nasdaq 100 ETF (-0.69 billion yuan)[6]

- The report mentions the construction of an "industry crowding monitoring model" to track the crowding levels of Shenwan first-level industry indices on a daily basis. The model identifies industries with high crowding levels, such as utilities and building materials, and those with lower levels, like automobiles and food & beverage. It also highlights significant daily changes in crowding levels for industries like real estate and utilities[6] - Another model mentioned is the "premium rate Z-score model," which is used to screen ETF products for potential arbitrage opportunities. The model employs rolling calculations to identify ETFs with potential risks of price corrections[6] - The industry crowding monitoring model evaluates crowding levels based on daily fund flows and crowding metrics, providing insights into industry trends and fund allocation changes over recent trading days[6] - The premium rate Z-score model calculates Z-scores for ETF premium rates, identifying deviations from historical averages that may signal arbitrage opportunities or risks[6] - The industry crowding monitoring model is qualitatively assessed as effective for identifying industry trends and fund allocation shifts, aiding investors in decision-making[6] - The premium rate Z-score model is qualitatively evaluated as useful for detecting arbitrage opportunities and potential risks in ETF pricing[6] - The industry crowding monitoring model highlights utilities and building materials as having high crowding levels, while automobiles and food & beverage exhibit lower levels. Real estate and utilities show significant daily crowding level changes[6] - The premium rate Z-score model identifies ETFs with potential arbitrage opportunities based on deviations in premium rates, though specific Z-score values are not provided in the report[6]

- The report uses Mahalanobis distance to measure industry crowding based on weekly price and transaction volume changes[3][17][18] - The construction process involves identifying industries where price and transaction volume deviate significantly, with industries outside the ellipse in quadrant 1 indicating short-term significant crowding[17] - Last week, the building materials industry showed significant trading crowding[18]

Core Viewpoint - The market is expected to continue its upward trend in the coming week, supported by various technical and macroeconomic indicators [1][2]. Market Indicators - The liquidity shock indicator for the CSI 300 index was 1.36, indicating current market liquidity is 1.36 times higher than the average level over the past year [2]. - The PUT-CALL ratio for the SSE 50 ETF decreased to 0.95, suggesting reduced caution among investors regarding short-term movements [2]. - The five-day average turnover rates for the SSE Composite Index and Wind All A were 0.99% and 1.63%, respectively, indicating increased trading activity [2]. Macroeconomic Factors - The RMB exchange rate fluctuated, with onshore and offshore rates increasing by 0.2% and 0.09% respectively [2]. - Historical data shows that from 2005 onwards, the probability of the SSE Composite Index, CSI 300, CSI 500, and ChiNext Index rising in the first half of July is 60%, 60%, 55%, and 53%, with average gains of 0.67%, 0.93%, 1.55%, and 1.6% respectively [2]. Event-Driven Insights - The US stock market rebounded, with the Dow Jones, S&P 500, and Nasdaq indices posting weekly returns of 3.82%, 3.44%, and 4.25% respectively [2]. - Several Federal Reserve officials signaled a dovish stance, with discussions around potential interest rate cuts in July if inflation remains controlled [2]. Technical Analysis - The Wind All A index broke above the SAR point on June 24, generating a buy signal [2]. - The current market score based on the moving average strength index is 216, placing it in the 85.1% percentile since 2021 [2]. - The sentiment model score is 3 out of 5, indicating a positive trend and sentiment in the market [2]. Market Performance - For the week of June 23-27, the SSE 50 index rose by 1.27%, the CSI 300 index by 1.95%, the CSI 500 index by 3.98%, and the ChiNext index by 5.69% [3]. - The overall market PE (TTM) stands at 19.7 times, which is in the 57.5% percentile since 2005 [3]. Factor Observations - The crowding degree for small-cap factors continues to decline, with a score of 0.74 for small-cap factors, -0.48 for low valuation factors, -0.31 for high profitability factors, and -0.15 for high growth factors [3]. - The industry crowding degree is relatively high in banking, non-ferrous metals, comprehensive, non-bank financials, and retail sectors, with significant increases in non-bank financials and banking [3].

Core Viewpoint - The market is expected to continue its oscillating trend in the upcoming week due to weak market sentiment and technical indicators suggesting a downward trend [1][2]. Market Overview - The liquidity shock indicator for the CSI 300 index was 1.23, indicating current market liquidity is 1.23 times higher than the average level over the past year [2]. - The PUT-CALL ratio for the SSE 50 ETF options increased to 1.06, reflecting a growing caution among investors regarding the short-term performance of the SSE 50 ETF [2]. - The five-day average turnover rates for the SSE Composite Index and Wind All A were 0.81% and 1.37%, respectively, indicating a decrease in trading activity [2]. Macroeconomic Factors - The onshore and offshore RMB exchange rates experienced slight fluctuations, with weekly changes of -0.03% and 0.14%, respectively [2]. - Recent economic data from the National Bureau of Statistics showed that in May, the industrial added value for large-scale enterprises grew by 5.8% year-on-year, and retail sales of consumer goods increased by 6.4% [2]. - Fixed asset investment for the first five months of the year rose by 3.7% year-on-year, with high-tech manufacturing and digital economy sectors showing significant growth [2]. Technical Analysis - The Wind All A index broke below the SAR point on June 19, indicating a bearish trend [2]. - The market score based on the moving average strength index is currently at 102, which is at the 39.7% percentile since 2021 [2]. - The sentiment model scored 1 out of 5, indicating weak market sentiment, while the trend model signal is positive and the weighted model signal is negative [2]. Market Performance - For the week of June 16-20, the SSE 50 index fell by 0.1%, the CSI 300 index decreased by 0.45%, the CSI 500 index dropped by 1.75%, and the ChiNext index declined by 1.66% [3]. - The overall market PE (TTM) stands at 19.2 times, which is at the 52.3% percentile since 2005 [3]. Factor Observations - The crowding degree for small-cap factors has decreased, with a current score of 0.79 for small-cap factors, -0.14 for low valuation factors, -0.11 for high profitability factors, and 0.00 for high profitability growth factors [3]. - The industry crowding degree is relatively high for sectors such as comprehensive, environmental protection, machinery equipment, banking, and non-ferrous metals, with notable increases in banking and medical biotechnology sectors [3].