债券研究 证 券 研 究 报 告 【债券周报】 2025H2 转债强赎与不强赎的变化 ——可转债周报 20250915 不强赎承诺期时长回落稳定，结构上以 1-3 个月内的为主。从 2020 年下旬开 始发行人设置不强赎承诺期时长基本呈稳定增长的形态，直至 2024 年初达到 最高点：承诺期普遍在 200 天左右。但 2024 年下半年至今不强赎承诺期回落 至 100-150 天左右实现稳定。究其原因：一方面，早期由于信披规则不成熟， 部分公司选择反复做本次不强赎或者一个月内不强赎的公告，所以使得均值不 高，后续 3 个月或 6 个月及以上的承诺期逐渐成为市场主流。另一方面，由于 2023 年后较长时间权益市场走势偏弱，使得触发强赎的转债也偏少，公告不 强赎的转债基本为长期不考虑的发行人故在短期内将承诺期拉长。 上周爱科科技、中仑新材、南芯科技新增董事会预案、5 家新增股东大会通过、 3 家新增发审委审批通过、2 家新增证监会核准，较去年同期分别+3、+5、+2、 +1。 承诺期长短对转债估值影响不大。2025 年初以来不强赎公告次日的转股溢价 率从 2%一直抬升至 8%左右，但较 2022 年的高位仍有部分 ...

Investment Rating - The report maintains a "Recommended" investment rating for China Overseas Development (00688.HK) with a target price of HKD 20 [1][8]. Core Views - The company achieved a revenue of HKD 832 billion in the first half of 2025, a year-on-year decrease of 4.3%, and a net profit attributable to ordinary shareholders of HKD 86 billion, down 16.6% year-on-year [1][7]. - The "Good House" system has established a market benchmark, and the company is actively replenishing its land reserves, with an investment intensity of 33.4% in the first half of 2025 [7][8]. - The company continues to focus on first-tier cities, with a sales area of 5.12 million square meters, a decline of 5.9% year-on-year, and a sales amount of HKD 120.2 billion, down 19.0% year-on-year [7][8]. Financial Performance Summary - The company's gross profit margin for the first half of 2025 was 17.4%, with a core net profit margin of 10.6% [7]. - The total land reserve area as of the end of June 2025 was 26.93 million square meters, with an equity area of 23.67 million square meters [7]. - The company reported a commercial operation income of HKD 3.54 billion, with shopping centers and office buildings contributing 81% of the revenue [7]. Debt and Cash Flow Summary - The company's interest-bearing debt decreased to HKD 227.5 billion, down HKD 14.1 billion, with a debt-to-asset ratio of 53.7% [7]. - The average financing cost for the first half of 2025 was 2.9%, and the operating cash flow remained positive [7][8]. - The report forecasts EPS for 2025-2027 to be HKD 1.40, HKD 1.38, and HKD 1.52 respectively, with a projected market capitalization of approximately HKD 218.8 billion [7][8].

Group 1 - The report maintains a positive outlook for the short term, indicating that it is not yet time for high-low switching, while mid-term expectations are for a physical re-inflation bull market [3][10][14] - The report emphasizes the importance of technology innovation, highlighting that the technology sector is expected to continue its growth, particularly in industries with clear growth expectations such as pharmaceuticals (innovative drugs), electronics (PCB), and communications (optical modules) [6][54] - The report notes a shift in market dynamics, with large-cap stocks outperforming small-cap stocks, driven by factors such as superior earnings under inflation, resilience in return on equity (ROE), and the expansion of ETFs favoring large-cap styles [12][34][35] Group 2 - The report discusses the "barbell strategy," which is suitable for low-price environments, indicating that as inflation expectations rise, the demand for the reverse barbell strategy will increase [4][19] - The report highlights the performance of the technology bull market and the return of leading blue-chip stocks, noting that since June 25, there has been a reversal in style within the technology sector, with large-cap stocks gaining significant traction [5][33][36] - The report identifies key industries to focus on in the mid-term, particularly those experiencing supply constraints and price increases due to the ongoing "anti-involution" policies, including industrial metals, small metals, steel, petrochemicals, and construction materials [6][56]

Investment Rating - The report maintains a positive investment rating for the automotive industry, particularly highlighting the performance of the robotics supply chain and suggesting a focus on leading companies and new opportunities [3]. Core Insights - The automotive sector's performance is primarily driven by the robotics supply chain, with catalysts expected to persist in the second half of the year. Traditional supply chains remain weak. The Ministry of Industry and Information Technology's "Automotive Industry Stabilization Growth Work Plan (2025-2026)" emphasizes the industrialization of intelligent connected technologies, including the conditional approval for L3 level vehicle production and accelerating breakthroughs in key technologies such as automotive chips, operating systems, artificial intelligence, and solid-state batteries. This indicates a shift in industry support from total volume logic to new technology logic [3][4]. Data Tracking - In August, new energy vehicle deliveries showed significant growth, with XPeng delivering 37,709 units, a year-on-year increase of 168.7%. BYD delivered 373,626 units, up 0.1% year-on-year, while traditional automakers like SAIC and Geely also reported substantial increases in sales [5][23][26]. - The average discount rate in late August was 9.9%, a decrease of 0.2 percentage points from earlier in the month, while the average discount amount was 22,198 yuan, down 344 yuan [5][27][28]. - The report recommends continued focus on electric vehicle manufacturers and traditional automakers with potential recovery, such as Jianghuai Automobile and SAIC [7]. Industry News - The report highlights several key developments, including the listing of Chery Automobile on the Hong Kong Stock Exchange and the launch of new models by various manufacturers, such as the LYNK & CO 10 EM-P and BYD's new SUV, the Titanium 7 [32][33]. - The retail sales of passenger vehicles in August reached 1.995 million units, a year-on-year increase of 4.6%, with new energy vehicles accounting for 110,100 units sold, up 7.5% year-on-year [32]. Market Performance - The automotive sector saw a weekly increase of 1.46%, ranking 17th out of 29 sectors. The overall market indices also showed positive growth, with the Shanghai Composite Index rising by 1.52% [10][36].

Quantitative Models and Construction Methods 1. Model Name: Volume Model - **Construction Idea**: The model uses trading volume data to predict market trends. - **Construction Process**: The model analyzes the trading volume of various broad-based indices to determine market sentiment. It categorizes the indices as neutral based on the volume data. - **Evaluation**: The model is considered neutral for all broad-based indices in the short term.[2][11] 2. Model Name: Low Volatility Model - **Construction Idea**: This model uses the volatility of stock prices to predict market trends. - **Construction Process**: The model evaluates the volatility of stock prices and categorizes the indices as neutral. - **Evaluation**: The model is considered neutral in the short term.[2][11] 3. Model Name: Institutional Feature Model - **Construction Idea**: This model uses institutional trading data from the "Dragon and Tiger List" to predict market trends. - **Construction Process**: The model analyzes the trading behavior of institutions listed on the "Dragon and Tiger List" and categorizes the indices as bullish. - **Evaluation**: The model is considered bullish in the short term.[2][11] 4. Model Name: Feature Volume Model - **Construction Idea**: This model uses specific volume features to predict market trends. - **Construction Process**: The model analyzes specific volume features and categorizes the indices as bearish. - **Evaluation**: The model is considered bearish in the short term.[2][11] 5. Model Name: Smart Algorithm Model (CSI 300) - **Construction Idea**: This model uses smart algorithms to predict market trends for the CSI 300 index. - **Construction Process**: The model applies smart algorithms to the CSI 300 index and categorizes it as neutral. - **Evaluation**: The model is considered neutral in the short term.[2][11] 6. Model Name: Smart Algorithm Model (CSI 500) - **Construction Idea**: This model uses smart algorithms to predict market trends for the CSI 500 index. - **Construction Process**: The model applies smart algorithms to the CSI 500 index and categorizes it as bearish. - **Evaluation**: The model is considered bearish in the short term.[2][11] 7. Model Name: Limit Up/Down Model - **Construction Idea**: This model uses the occurrence of limit up and limit down events to predict market trends. - **Construction Process**: The model analyzes the frequency of limit up and limit down events and categorizes the indices as neutral. - **Evaluation**: The model is considered neutral in the medium term.[2][12] 8. Model Name: Calendar Effect Model - **Construction Idea**: This model uses calendar effects to predict market trends. - **Construction Process**: The model analyzes historical calendar effects and categorizes the indices as neutral. - **Evaluation**: The model is considered neutral in the medium term.[2][12] 9. Model Name: Long-term Momentum Model - **Construction Idea**: This model uses long-term momentum to predict market trends. - **Construction Process**: The model analyzes long-term momentum indicators and categorizes the indices as bullish. - **Evaluation**: The model is considered bullish in the long term.[2][13] 10. Model Name: Comprehensive Weapon V3 Model - **Construction Idea**: This model combines multiple factors to predict market trends. - **Construction Process**: The model integrates various factors and categorizes the indices as bearish. - **Evaluation**: The model is considered bearish in the long term.[2][14] 11. Model Name: Comprehensive National Certificate 2000 Model - **Construction Idea**: This model combines multiple factors to predict market trends for the National Certificate 2000 index. - **Construction Process**: The model integrates various factors and categorizes the indices as bearish. - **Evaluation**: The model is considered bearish in the long term.[2][14] 12. Model Name: Turnover Inverse Amplitude Model - **Construction Idea**: This model uses the inverse amplitude of turnover to predict market trends. - **Construction Process**: The model analyzes the inverse amplitude of turnover and categorizes the indices as bullish. - **Evaluation**: The model is considered bullish in the medium term.[2][15] Model Backtest Results - **Volume Model**: Neutral for all broad-based indices in the short term.[2][11] - **Low Volatility Model**: Neutral in the short term.[2][11] - **Institutional Feature Model**: Bullish in the short term.[2][11] - **Feature Volume Model**: Bearish in the short term.[2][11] - **Smart Algorithm Model (CSI 300)**: Neutral in the short term.[2][11] - **Smart Algorithm Model (CSI 500)**: Bearish in the short term.[2][11] - **Limit Up/Down Model**: Neutral in the medium term.[2][12] - **Calendar Effect Model**: Neutral in the medium term.[2][12] - **Long-term Momentum Model**: Bullish in the long term.[2][13] - **Comprehensive Weapon V3 Model**: Bearish in the long term.[2][14] - **Comprehensive National Certificate 2000 Model**: Bearish in the long term.[2][14] - **Turnover Inverse Amplitude Model**: Bullish in the medium term.[2][15]

- The report introduces a "Total Heat Indicator" to monitor market sentiment, defined as the sum of browsing, self-selection, and click counts for individual stocks, normalized by their market share on the same day, and multiplied by 10,000, with a range of [0,10000][7] - A simple rotation strategy is constructed based on weekly heat change rates (MA2), where the highest heat change rate group is selected for investment at the end of each week, and if the "Other" group has the highest rate, the portfolio remains empty[12][15] - The rotation strategy achieves an annualized return of 8.74% since 2017, with a maximum drawdown of 23.5%, and a return of 32.7% in 2025[15] - A concept-based strategy is developed by selecting the top 5 concepts with the highest heat change rates weekly, forming two portfolios: "TOP" (top 10 stocks with the highest heat within each concept) and "BOTTOM" (bottom 10 stocks with the lowest heat within each concept)[30] - The "BOTTOM" portfolio historically delivers an annualized return of 15.71%, with a maximum drawdown of 28.89%, and a return of 44.2% in 2025[32]

Report Summary 1) Report Industry Investment Rating No information provided on the industry investment rating in the report. 2) Core Viewpoints - The convertible bond market showed positive trends on September 12, 2025, with more than half of the industries rising and the valuation increasing. The trading sentiment in the convertible bond market also heated up, and small - cap growth stocks were relatively dominant [1]. - The central price of convertible bonds increased, while the proportion of high - price bonds decreased. The overall valuation of convertible bonds rose, with the conversion premium rate of various types of convertible bonds increasing [2]. 3) Summary by Related Catalogs Market Overview - Index Performance: The CSI Convertible Bond Index rose 0.17% month - on - month, while the Shanghai Composite Index, Shenzhen Component Index, ChiNext Index, and Shanghai 50 Index all declined. The CSI 1000 Index rose 0.31% [1]. - Market Style: Small - cap growth stocks were relatively dominant. Among them, small - cap growth stocks rose 0.32%, while large - cap growth and value stocks declined [1]. - Capital Performance: The trading volume in the convertible bond market reached 82.886 billion yuan, a 7.82% increase month - on - month. The total trading volume of the Wind All - A Index was 2.548312 trillion yuan, a 3.40% increase. The net outflow of main funds from the Shanghai and Shenzhen stock markets was 37.278 billion yuan, and the yield of the 10 - year Treasury bond decreased by 0.73bp to 1.87% [1]. Convertible Bond Price and Valuation - Price: The weighted average closing price of convertible bonds was 131.41 yuan, a 0.17% increase. The proportion of high - price bonds (above 130 yuan) decreased by 0.29pct, and the proportion of bonds in the 110 - 120 yuan range increased by 1.0pct. There were no bonds with a closing price below 100 yuan [2]. - Valuation: The conversion premium rate of 100 - yuan par - value convertible bonds was 29.94%, a 0.38pct increase. The overall weighted par value decreased by 0.52%. The conversion premium rates of all types of convertible bonds (including partial - equity, partial - debt, and balanced) increased [2]. Industry Performance - Stock Market: Among A - share industries, the top three decliners were communication (-2.13%), beauty care (-1.52%), and banking (-1.52%); the top three gainers were non - ferrous metals (+1.96%), real estate (+1.51%), and steel (+1.41%) [3]. - Convertible Bond Market: 18 industries in the convertible bond market rose. The top three gainers were environmental protection (+2.86%), non - ferrous metals (+1.51%), and communication (+1.37%); the top three decliners were machinery and equipment (-2.08%), media (-1.56%), and agriculture, forestry, animal husbandry, and fishery (-0.48%) [3]. - Different Industry Indicators: In terms of closing price, large - cycle industries rose 0.89%, while manufacturing industries declined 0.22%. In terms of conversion premium rate, all industries increased to varying degrees. In terms of conversion value, large - cycle industries rose 0.87%, while manufacturing industries declined 1.60% [3].

Investment Rating - The report does not explicitly provide an investment rating for the humanoid robotics industry Core Insights - The humanoid robotics industry is in the early stages of transitioning from laboratory to commercial applications, with significant potential driven by leaders like Tesla's Optimus pushing for large-scale deployment [2][5] - The industry faces five key conditions for full commercialization: technological maturity, cost control, clear market positioning, infrastructure and ecosystem development, and social/legal acceptance [2][5] - The current core contradiction lies between high costs and unproven application scenarios/ROI, which constrains market demand and cost reduction efforts [2][5] Summary by Sections Upstream Core Components - The upstream segment is characterized by high technical barriers and cost concentration, determining the performance limits of robots, including components like high-torque density joint assemblies and AI chips [8][9] - The value of the three main actuators constitutes 73% of the humanoid robot's main component value, with high precision and performance requirements [9][12] - Domestic suppliers have significant room for improvement in high-tech components, with some achieving breakthroughs in key areas like harmonic reducers [12][17] Midstream Body Design, Manufacturing, and Integration - The midstream focuses on assembling core components into complete robotic products, requiring strong integration capabilities and scalable production [36][39] - Key tasks include body design, assembly manufacturing, system integration, and testing to ensure compatibility and performance [37][38] - The competition in the midstream is intense, with various companies attempting to overcome technological and commercial barriers [39] Downstream Applications and Services - The downstream segment has diverse application scenarios, ranging from industrial to personal services, with significant growth potential [2][5] - The commercialization of humanoid robots requires meeting specific conditions, including technological maturity and market demand alignment [2][5] - The market space is expected to expand exponentially with technological breakthroughs and demand surges [2][5] Potential Business Model Exploration - Future winners in the humanoid robotics industry will be those who can create integrated hardware, software, and ecosystem services [2][5] - Short-term opportunities may lie in B2B RaaS models and specific industry solutions, while long-term value will come from software platforms and ecosystems [2][5]

Investment Rating - The report maintains a "Strong Buy" rating for the company, expecting it to outperform the benchmark index by over 20% in the next six months [2][19]. Core Insights - The company's profits are under pressure due to price impacts, but high-end lithium iron phosphate (LiFePO4) research and development is progressing smoothly [2]. - The company reported a revenue of 3.882 billion yuan for H1 2025, a year-on-year decrease of 10.58%, with a net loss attributable to shareholders of 391 million yuan, an improvement from a loss of 516 million yuan in the same period last year [8]. - The company is increasing its R&D investment, with a total of 102 million yuan spent and 61 new patent applications filed during the reporting period [8]. Financial Performance Summary - Total revenue is projected to increase from 7.613 billion yuan in 2024 to 14.359 billion yuan in 2027, with a compound annual growth rate (CAGR) of 19.6% [4]. - The net profit attributable to shareholders is expected to improve from a loss of 1.338 billion yuan in 2024 to a profit of 434 million yuan in 2027, reflecting a significant turnaround [4]. - Earnings per share (EPS) is forecasted to rise from -4.77 yuan in 2024 to 1.55 yuan in 2027 [4]. Product Development and Market Position - The company has the largest established production capacity for manganese iron lithium at 110,000 tons per year, with products already in mass production [8]. - The first-generation manganese iron lithium product has achieved batch delivery, and the second-generation product is progressing well in validation [8]. - The company’s lithium supplement enhancer is the first in the world to achieve mass production, with a capacity of 5,000 tons per year, leading the industry in both scale and progress [8].

Investment Rating - The report maintains a "Recommendation" rating for the company, indicating an expectation to outperform the benchmark index by 10%-20% over the next six months [1][17]. Core Insights - The company's operating performance aligns with its earnings forecast, with a focus on the progress of the highly anticipated new product "Yihuan" [1]. - The company achieved a revenue of 3.691 billion yuan in H1 2025, a year-on-year increase of 33.74%, and turned a profit with a net profit of 503 million yuan [7]. - The gaming segment contributed significantly to revenue growth, driven by the self-developed MMORPG "Zhuxian World" and continued growth in esports titles like "CS:GO" and "DOTA2" [7]. - The film and television segment saw a remarkable revenue increase of 756.35% year-on-year, with several popular series released in 2025 [7]. - The upcoming game "Yihuan" is expected to enhance the company's market position and revenue potential, with ongoing adjustments based on player feedback [7]. Financial Summary - Total revenue projections for 2025-2027 have been adjusted to 6.5 billion, 10.2 billion, and 9.9 billion yuan respectively, reflecting year-on-year growth rates of 16%, 57%, and -3% [7]. - The forecast for net profit has been revised to 700 million, 2.1 billion, and 2 billion yuan for the same period, indicating significant recovery and growth [7]. - The target price for the company's stock is set at 21.65 yuan, based on a projected price-to-earnings ratio of 20x for 2026 [7].