- Model Name: "Day and Night Convergence" New Momentum Factor; Model Construction Idea: Based on the price-volume relationship during the day and overnight, the intraday factor and overnight factor are improved and then recombined into a new momentum factor[1][7] - Model Construction Process: 1. Split the trading period into day and night sessions[7] 2. Explore the price-volume relationship in each session separately[7] 3. Construct the "Day and Night Convergence" new momentum factor based on the findings[7] - Model Evaluation: The new factor significantly outperforms the traditional momentum factor in terms of stock selection ability[6][7] - Factor Name: "Day and Night Convergence" New Momentum Factor; Factor Construction Idea: Incorporate the information of "trading volume" into the previous "momentum factor segmentation" research to further explore the differences in investor trading behavior[7] - Factor Construction Process: 1. Split the trading period into day and night sessions[7] 2. Explore the price-volume relationship in each session separately[7] 3. Construct the "Day and Night Convergence" new momentum factor based on the findings[7] - Factor Evaluation: The new factor significantly outperforms the traditional momentum factor in terms of stock selection ability[6][7] Model Backtest Results - "Day and Night Convergence" New Momentum Factor, Annualized Return: 17.95%, Annualized Volatility: 8.70%, IR: 2.06, Monthly Win Rate: 77.86%, Maximum Drawdown: 9.07%[1][7][14] Factor Backtest Results - "Day and Night Convergence" New Momentum Factor, Annualized Return: 17.95%, Annualized Volatility: 8.70%, IR: 2.06, Monthly Win Rate: 77.86%, Maximum Drawdown: 9.07%[1][7][14]

Quantitative Models and Construction Methods 1. **Model Name**: "Regain Confidence 2.0" RCP Factor - **Model Construction Idea**: The model is based on the behavioral finance concept of overconfidence. It innovatively uses high-frequency minute sequence data to calculate the time gap between positive news surges and stock price corrections to construct the overconfidence factor CP. The second-generation RCP factor is derived by orthogonalizing the first-generation CP factor with intraday returns, considering the potential overcorrection after overconfidence.[1][6] - **Model Construction Process**: - Calculate the time gap between positive news surges and stock price corrections to construct the overconfidence factor CP. - Orthogonalize the CP factor with intraday returns to obtain the residuals, which form the second-generation RCP factor. - Use standardized factors instead of ranking values to retain factor information, improving the purified effect of the new RCP factor.[6][7] - **Model Evaluation**: The RCP factor-based portfolio significantly outperforms traditional portfolio methods.[6] Model Backtesting Results 1. **"Regain Confidence 2.0" RCP Factor**: - Annualized Return: 17.66%[1][7][10] - Annualized Volatility: 7.87%[1][7][10] - Information Ratio (IR): 2.24[1][7][10] - Monthly Win Rate: 77.14%[1][7][10] - Maximum Drawdown: 7.46%[1][7][10] - September Performance: Long portfolio return 1.00%, short portfolio return -0.97%, long-short hedged return 1.97%[1][10]

Domestic Macro Viewpoints - The impact of the new round of tariffs on the domestic economy is expected to be limited, with a 16.9% year-on-year decline in exports to the U.S. in the first nine months, yet overall exports still achieved a 6.1% year-on-year growth[3] - Since Q3, domestic economic pressure has increased, with fixed asset investment growth dropping to 0.5% year-on-year in August and retail sales growth at 3.4% year-on-year, indicating a need for new growth stabilization policies[3] - The anticipated new growth stabilization policies are expected to be moderate, focusing on support rather than strong stimulus, with Q3 economic growth projected between 4.7% and 4.9%[3] U.S. Economic Outlook - The U.S. economy is expected to remain resilient, with the Federal Reserve likely to implement two more rate cuts, although the market has already priced in approximately 4.7 rate cuts by the end of next year, limiting further rate reduction space[3] - Market sentiment regarding tariffs is divided, with optimistic views suggesting a quick rebound in U.S. and Chinese stock markets, while pessimistic views foresee potential corrections due to a lack of substantial concessions[3] Market Trends - Following the tariff-related adjustments, the market is expected to enter a consolidation phase from October to November, with a potential shift from AI hardware to defensive sectors and industries supported by performance logic[3] - The bond market is experiencing a temporary downward adjustment in rates, with the 10-year yield expected to stabilize between 1.70% and 1.75% due to external risks and domestic economic fundamentals[4]

Investment Rating - The investment rating for the company is "Buy" [1] Core Views - The company is expected to benefit from increased investment income, with a projected year-on-year growth in net profit attributable to shareholders for Q3 ranging from 58% to 101% [1] - The report anticipates a significant increase in net profit for the first three quarters of 2025, estimated between 30 billion to 34.1 billion yuan, reflecting a year-on-year growth of 45% to 65% [1] - The report highlights that the growth in net profit is primarily driven by improved investment returns due to a recovering domestic capital market [1] Financial Forecasts - Revenue projections for the company are as follows: - 2023: 71,547 million yuan - 2024: 132,555 million yuan (up 85.3% YoY) - 2025: 148,303 million yuan (up 11.9% YoY) - 2026: 154,262 million yuan (up 4.0% YoY) - 2027: 160,653 million yuan (up 4.1% YoY) [1] - Net profit attributable to shareholders is forecasted as: - 2023: 8,712 million yuan - 2024: 26,229 million yuan (up 201.1% YoY) - 2025: 38,702 million yuan (up 47.6% YoY) - 2026: 40,210 million yuan (up 3.9% YoY) - 2027: 41,832 million yuan (up 4.0% YoY) [1] - The estimated enterprise value (EV) per share is projected to increase from 80.30 yuan in 2023 to 106.88 yuan in 2027 [1] Market Data - The closing price of the company's stock is 62.18 yuan, with a market capitalization of approximately 193,973.41 million yuan [5] - The price-to-book ratio (P/B) is reported at 2.33 [5] - The company has a total share capital of approximately 3,119.55 million shares, with 2,085.44 million shares in circulation [6]

- The report analyzes the investment value of the Guotai SSE 10-Year Treasury Bond ETF, highlighting its advantages in terms of low fee rates, high transparency, and efficient tracking of the SSE 10-Year Treasury Bond Index[4][8][50] - The SSE 10-Year Treasury Bond Index (code: H11077.SH) is a bond index launched by the Shanghai Stock Exchange on March 7, 2013. It is composed of treasury bonds with remaining maturities between 6.5 and 10.25 years, calculated using a market capitalization-weighted method to reflect the overall price trend of treasury bonds in this maturity range[45][46][47] - The Guotai SSE 10-Year Treasury Bond ETF tracks the SSE 10-Year Treasury Bond Index, investing at least 90% of its net assets in the index's constituent bonds and alternative constituent bonds. The ETF aims to replicate the index's performance with minimal tracking error, providing investors with a convenient way to access a basket of high-credit-quality, liquid medium- to long-term treasury bonds[50][51][54] - The ETF demonstrates strong performance metrics: annualized return of 3.81%, annualized volatility of 2.65%, IR of 1.44, monthly win rate of 71.13%, and maximum drawdown of 3.79%. Relative to its benchmark, it achieves an annualized excess return of 2.20%, excess volatility of 0.59%, excess IR of 3.72, excess monthly win rate of 93.81%, and excess maximum drawdown of 0.73%[59][63][62] - The ETF's historical excess performance is consistently positive, with monthly excess win rates of 100% since 2021 and zero monthly excess drawdowns during the same period. For example, in 2021, the excess IR reached 13.42, and in 2023, it further improved to 21.22[63][62][59]

Macro Strategy - The report indicates that the economic situation is influenced by the upcoming measures for counter-cyclical adjustments, with a focus on the timing of restarting government bond trading and the potential for interest rate cuts depending on the economic outlook for Q4 2025 [1][12][13] - The ECI supply index is at 49.99%, showing a slight decline, while the demand index is at 49.91%, indicating a mixed economic recovery influenced by the recent holidays [12][13] - The report highlights the potential limited impact of Trump's renewed tariff threats on the US economy, while emphasizing the need to monitor retaliatory measures and the escalation of trade conflicts into critical sectors like rare earths and semiconductors [1][12][14] Fixed Income - The report discusses the recent issuance of secondary capital bonds, with one bond issued at a scale of 1.2 billion yuan, and a total trading volume of approximately 44.5 billion yuan, which is a significant decrease from the previous week [3][20] - It notes the issuance of green bonds totaling approximately 15.25 billion yuan, with a decrease in trading volume to 41.8 billion yuan, indicating a cooling in the market [4][21] - The report suggests that investors should prioritize controllable pullbacks in convertible bonds and focus on performance improvement or valuation recovery rather than getting caught up in clause negotiations [15][18] Company Analysis - Hong Kong Travel (00308.HK) is focusing on core profitable businesses by divesting its tourism real estate assets, with projected net profits of 270 million, 420 million, and 600 million HKD for 2025-2027, corresponding to PE ratios of 31, 20, and 14 times [5][7] - Guangyang Co., Ltd. (002708) is expected to achieve revenues of 2.774 billion, 3.700 billion, and 4.795 billion yuan, with net profits of 108 million, 218 million, and 363 million yuan for 2025-2027, indicating a recovery in profitability and new growth opportunities in FPC, low-altitude economy, and humanoid robots [8] - Hengyi Petrochemical (000703) anticipates net profits of 430 million, 650 million, and 820 million yuan for 2025-2027, benefiting from the recovery of the polyester industry and the upcoming production of its Qinzhou project [9] - Rongsheng Petrochemical (002493) is projected to have net profits of 1.9 billion, 2.9 billion, and 4.1 billion yuan for 2025-2027, with a focus on benefiting from stable growth policies in the petrochemical sector [11]

- The style rotation model is constructed based on the Dongwu quantitative multi-factor system, starting from micro-level stock factors. It selects 80 underlying factors as original features, including valuation, market capitalization, volatility, and momentum, and further constructs 640 micro features. The model replaces the absolute proportion division of style factors with common indices as style stock pools, creating new style returns as labels. A random forest model is trained in a rolling manner to avoid overfitting risks, optimizing features and obtaining style recommendations. The framework integrates style timing, scoring, and actual investment[9][4] - The performance of the style rotation model during the backtesting period (2017/01/01-2025/09/30) shows an annualized return of 16.41%, annualized volatility of 20.43%, IR of 0.80, monthly win rate of 58.49%, and a maximum drawdown of 25.54%. When hedging against the market benchmark, the annualized return is 10.54%, annualized volatility is 10.85%, IR is 0.97, monthly win rate is 55.66%, and the maximum drawdown is 8.79%[10][11] - The style rotation model's latest timing directions for October 2025 are value, large market capitalization, momentum, and low volatility[2][19] - The latest holdings of the style rotation model for October 2025 include indices such as CSI Central Enterprise Dividend (ETF code: 561580.SH), CSI Bank (ETF code: 512700.SH), CSI Film and Television (ETF code: 159855.SZ), CS Battery (ETF code: 159796.SZ), and CSI All Real Estate (ETF code: 512200.SH)[3][19]

Investment Rating - The report maintains an "Accumulate" rating for the environmental protection industry [1] Core Views - The report highlights the benefits of rising metal prices for companies like High Energy Environment and the performance contributions from green electricity in mining for Longjing Environmental Protection. It emphasizes the value of water and solid waste dividends [1] Industry Trends - The environmental protection industry is expected to see a significant increase in cash flow and dividends due to reduced capital expenditures and improved operational efficiency. The solid waste sector is entering a mature phase, with free cash flow turning positive in 2023 and continuing to improve in 2024 [18][20] - The report notes that the market for water services is stabilizing, with a focus on cash flow improvements and potential for high dividends, similar to the garbage incineration sector [23][24] Company Tracking - High Energy Environment is benefiting from rising metal prices and is actively expanding its resource recycling operations. Longjing Environmental Protection is seeing performance contributions from new projects in green electricity and storage equipment [5][18] - The report recommends several companies for investment, including Huanlan Environment, Green Power, and Yongxing Co., highlighting their strong dividend potential and operational improvements [5][23] Policy Tracking - The Ministry of Ecology and Environment is set to release a carbon emissions trading market allocation plan for the steel, cement, and aluminum industries, which will cover approximately 1,500 new key emission units and manage a total emission volume of 3 billion tons of CO2 equivalent [10][11]

- The five-dimensional industry rotation model is based on the unique multi-factor stock selection system of Dongwu Securities, starting from the micro-level of individual stocks. The micro-factors are categorized into five major types: volatility, fundamentals, trading volume, sentiment, and momentum. The model constructs intra-industry dispersion indicators and intra-industry traction indicators based on the style preferences of the large-category stock selection factors, ultimately obtaining five types of composite industry factors. The model includes volatility, fundamentals, trading volume, sentiment, and momentum dimensions[8][3] - The performance tracking of the five-dimensional industry rotation model includes the construction of return capability scores and stability scores based on the excess returns of the long group and the average industry ranking of the long group, using the period from April 1, 2025, to September 30, 2025, as the look-back period[9] - The backtesting performance of the five-dimensional industry rotation model from January 1, 2015, to September 30, 2025, shows that the annualized return of the six-group long-short hedging in the Shenwan primary industry is 21.10%, the annualized volatility is 10.84%, the information ratio (IR) is 1.95, the monthly win rate is 72.36%, and the maximum historical drawdown is 13.30%[12][13] - The long-short hedging performance indicators of the five-dimensional industry rotation model from January 1, 2015, to September 30, 2025, for each factor are as follows: Volatility factor: annualized return 10.48%, volatility 10.41%, IR 1.01, win rate 59.38%, maximum drawdown 14.81%; Fundamentals factor: annualized return 7.04%, volatility 12.12%, IR 0.58, win rate 56.25%, maximum drawdown 26.32%; Trading volume factor: annualized return 8.03%, volatility 11.78%, IR 0.68, win rate 59.38%, maximum drawdown 18.40%; Sentiment factor: annualized return 8.24%, volatility 12.79%, IR 0.64, win rate 64.84%, maximum drawdown 14.79%; Momentum factor: annualized return 11.50%, volatility 10.59%, IR 1.09, win rate 61.42%, maximum drawdown 13.52%; Composite factor: annualized return 21.10%, volatility 10.84%, IR 1.95, win rate 72.36%, maximum drawdown 13.30%[14] - The long-short hedging performance indicators of the five-dimensional industry rotation model from January 1, 2015, to September 30, 2025, for the long group hedging the equal-weighted industry portfolio are as follows: annualized return 10.36%, annualized volatility 6.57%, IR 1.58, monthly win rate 69.92%, maximum drawdown 9.36%[13][16] - In September 2025, the long group of the five-dimensional industry rotation model achieved a return of 2.00%, with an excess return relative to the benchmark of 1.53%[19] - The long-short return decomposition for September 2025 for each factor is as follows: Volatility factor: long -3.72%, short 4.21%; Fundamentals factor: long 2.10%, short -2.56%; Trading volume factor: long 0.19%, short 1.57%; Sentiment factor: long 4.89%, short 1.44%; Momentum factor: long -1.95%, short -2.43%; Composite factor: long 2.00%, short -0.28%[20]

- Model Name: Expected High Dividend Portfolio; Model Construction Idea: The model uses a two-stage approach to construct the expected dividend yield indicator. The first stage calculates the dividend yield based on the annual report's profit distribution, and the second stage predicts and calculates the dividend yield using historical dividends and fundamental indicators. Additionally, two short-term factors affecting dividend yield—reversal factor and profitability factor—are used to assist in screening, and the selection is made from the CSI 300 constituent stocks to construct the expected high dividend portfolio. The portfolio holds 30 stocks each period and rebalances monthly[3][8] - Model Construction Process: 1. Exclude suspended and limit-up stocks from the CSI 300 constituent stocks to form the candidate stock pool[13] 2. Exclude the top 20% of stocks with the highest short-term momentum (i.e., the top 20% of stocks with the highest 21-day cumulative gains) from the stock pool[13] 3. Exclude stocks with declining profitability (i.e., stocks with a negative year-on-year growth rate of quarterly net profit)[13] 4. Rank the remaining stocks in the stock pool by expected dividend yield and select the top 30 stocks with the highest expected dividend yield to construct the portfolio equally weighted[9] - Model Evaluation: The model's historical performance is outstanding, with a cumulative return of 358.90% and a cumulative excess return of 107.44% relative to the CSI 300 Total Return Index. The annualized excess return is 8.87%, with a maximum rolling one-year drawdown of only 12.26% and a monthly excess win rate of 60.19%[11] Model Backtest Results - Expected High Dividend Portfolio, average return in September 2025: -5.35%, underperformed the CSI 300 Index by 8.09% and the CSI Dividend Index by 3.87%[3][14] - Best performing stocks in September 2025: CITIC Special Steel (3.81%), Yutong Bus (-0.35%), Industrial and Commercial Bank of China (-1.75%), Shuanghui Development (-1.90%)[15] Factor Construction and Evaluation - Factor Name: Expected Dividend Yield Factor; Factor Construction Idea: The factor is constructed by predicting dividend distribution using the method of dividend distribution combined with fundamental indicators. Two short-term factors affecting dividend yield—reversal factor and profitability factor—are used to assist in screening[14] - Factor Construction Process: 1. Calculate the dividend yield based on the annual report's profit distribution[8] 2. Predict and calculate the dividend yield using historical dividends and fundamental indicators[8] - Factor Evaluation: The factor is used to assist in screening and constructing the expected high dividend portfolio, which has shown outstanding historical performance[3][8] Factor Backtest Results - Expected Dividend Yield Factor, average return in September 2025: -5.35%, underperformed the CSI 300 Index by 8.09% and the CSI Dividend Index by 3.87%[3][14]