[Table_IndNameRptType]2 固定收益 固收周报 超半数固收卖方看多债市 ——债市情绪面周报（6 月第 3 周） 报告日期： 2025-06-16 [Table_Author] 首席分析师：颜子琦 执业证书号：S0010522030002 电话：13127532070 邮箱：yanzq@hazq.com [Table_Author] 研究助理：洪子彦 执业证书号：S0010123060036 电话：15851599909 邮箱：hongziyan@hazq.com 主要观点： ⚫[Table_Summary] 华安观点：债市顺风，赔率有限，仍重交易 近期市场情绪升温，投资者看多也在做多，市场情绪（买方与卖方） 上升至年初至今最高位，同时投资者选择拉长久期与加杠杆，基本面依然 对债市有利（社融由政府债支撑、物价依然偏弱），大行买短债也成为趋 势，可关注后续宽货币的政策信号，但在债市顺风环境下，利率整体下行 的赔率有限，从曲线形态看长债较历史极值空间约为 3bp，同时跨半年环 境下拉长久期是机会也是风险，历史上 6 月债市整体涨跌参半，仍宜以交 易思维对待。 ⚫ 卖方观点：超半数固收卖方看多债市 ...

电力设备 行业周报 SNEC 展新品频出，海风项目稳步推进 行业评级：增持 报告日期： 2025-06-15 行业指数与沪深 300 走势比较 分析师：张志邦 执业证书号：S0010523120004 邮箱： zhangzhibang@hazq.com 分析师：刘千琳 执业证书号：S0010524050002 邮箱： liuqianlin@hazq.com 分析师：郑洋 执业证书号：S0010524110003 邮箱： zhengyang@hazq.com 相关报告 1.国内大储招标高增，关注虚拟电厂 环节 2025-06-09 2.绿电直连有望提振需求，海风项目 稳步推进 2025-06-05 3.海风基本面进入右侧，宝马全固态 电池汽车路测 2025-05-26 主要观点： -15% -10% -5% 0% 5% 10% 15% 20% 25% 06/13 07/13 08/13 09/13 10/13 11/13 12/13 01/13 02/13 03/13 04/13 05/13 06/13 电网设备 沪深300 近期，国家能源集团 2025 年第 1 批 4354.33MW 风力发电机组集采 公开 ...

[Table_IndNameRptType]2 固定收益 固收周报 债市投资者已从看多转向做多 ——债市机构行为周报（6 月第 3 周） 报告日期： 2025-06-15 [Table_Author] 首席分析师：颜子琦 执业证书号：S0010522030002 电话：13127532070 邮箱：yanzq@hazq.com [Table_Author] 研究助理：洪子彦 执业证书号：S0010123060036 电话：15851599909 邮箱：hongziyan@hazq.com ⚫[Table_Summary] 三个债市边际变化 也在拉长久期博资本利得。 第三，债市整体杠杆率也在回升，已超去年水平。一季度债市资金面整 体均衡偏紧，DR007 运行在政策利率上方，对应债市杠杆率持续处于低 位，远低于季节性水平。而近期我们看到，在央行提前公布买断式逆回 购+大行买短债的情况下，6 月资金面并不紧，这一现象也催化了机构加 杠杆的诉求。 综合以上三点，叠加近期披露的基本面数据依然对债市有利（社融由政 府债支撑、物价依然偏弱），债市看多+做多的债市行情已经来临。 ⚫ 同时也有三点关注 第一，拉久期+加杠杆环境 ...

Investment Rating - The industry investment rating is "Overweight" [1] Core Viewpoints - The report tracks the changes in operating income, net profit margin, and capital expenditure for the automotive parts sector in 2024A and Q1 2025, indicating that precision parts, chassis components, and electronic components show relatively good growth and capital expenditure intensity [4][10] - The report emphasizes the importance of capital expenditure for manufacturing enterprises to maintain operational capabilities, suggesting that capital expenditure is a precursor to future output and can help identify investment opportunities in sectors with high growth potential [4][10] Summary by Sections 1. Growth Potential - In 2024A, the operating income growth rates are as follows: electronic components (-), chassis components (+), body components (-), exterior components (-), overall industry (+), precision parts (+), powertrain components (+), interior components (+), and molds (-) [10][13] - The net profit margin for 2024 is 5.2%, a decrease of 0.6 percentage points compared to the previous year, while Q1 2025 shows a net profit margin of 5.8%, an increase of 0.6 percentage points [15][19] - The report categorizes the growth potential into four tiers based on operating income growth and net profit margin changes, with precision parts and chassis components performing well [10][11][15] 2. Capital Expenditure Intensity - The capital expenditure as a percentage of operating income for the automotive parts sector in 2024 is 6.9%, which is an increase compared to the previous year, while Q1 2025 shows a decrease to 6.4% [22] - The report identifies different tiers of capital expenditure intensity across various segments, with electronic components and molds showing high capital expenditure intensity [22] 3. Investment Recommendations - The report suggests focusing on sectors with high capital expenditure and growth potential, particularly precision parts (gears, bearings), chassis components (steering systems, suspension), electronic components, and thermal management systems [4][10] - Recommended companies include Shuanglin Co., Guansheng Co., Jifeng Co., and Huayang Group, with additional mentions of Zhejiang Shibao, Zhongding Co., Meili Technology, and others [4][10]

Quantitative Models and Construction Methods 1. Model Name: Optimal Factor Timing Portfolio - **Model Construction Idea**: The model integrates numerous factors and predictors to construct a timing strategy, leveraging shrinkage techniques to avoid overfitting to historical data and spurious opportunities[2][3][17] - **Model Construction Process**: 1. Use the Ledoit and Wolf (2003) covariance matrix shrinkage estimator to ensure robustness even with thousands of timing portfolios. The shrinkage intensity is calculated following Schäfer and Strimmer (2005)[3][25] 2. Apply a modified version of Kozak, Nagel, and Santosh (2020) shrinkage method to estimate portfolio weights, introducing skepticism toward unrealistically high Sharpe ratio opportunities. The weight formula is: $$ \hat{W}_{t}=\left(\hat{\Sigma}_{t}+\hat{\overline{t}}_{t}\left[\begin{array}{cc}0&0\\ 0&\hat{D}_{t}\end{array}\right]\right)^{-1}\hat{\mu}_{t} $$ where $\hat{\Sigma}_{t}$ is the covariance matrix, $\hat{\mu}_{t}$ is the mean return vector, and $\lambda$ controls the shrinkage degree[27][28][29] 3. Rescale portfolio weights to ensure the absolute sum of weights equals 1 in each period, focusing on factor rotation while avoiding extreme leverage[3][30] - **Model Evaluation**: The shrinkage techniques effectively mitigate estimation errors and prevent overfitting, ensuring robust out-of-sample performance even in high-dimensional settings[3][30][80] --- Model Backtesting Results 1. Optimal Factor Timing Portfolio (Fama-French Factors) - **Mean Return**: 4.71% - **Standard Deviation**: 5.81% - **Sharpe Ratio**: 0.81 - **Appraisal Ratio**: 0.79 - **Worst 12-Month Return**: -5.62%[40][41] 2. Optimal Factor Timing Portfolio (Large-Cap Fama-French Factors) - **Mean Return**: 3.65% - **Standard Deviation**: 6.49% - **Sharpe Ratio**: 0.56 - **Appraisal Ratio**: 0.54 - **Worst 12-Month Return**: -13.70%[40][41] 3. Optimal Factor Timing Portfolio (Jensen Factors, Small Predictor Set) - **Mean Return**: 2.97% - **Standard Deviation**: 2.01% - **Sharpe Ratio**: 1.48 - **Appraisal Ratio**: 1.51 - **Worst 12-Month Return**: -1.79%[40][41] 4. Optimal Factor Timing Portfolio (Jensen Factors, Large Predictor Set) - **Mean Return**: 2.73% - **Standard Deviation**: 1.91% - **Sharpe Ratio**: 1.43 - **Appraisal Ratio**: 1.46 - **Worst 12-Month Return**: -3.17%[40][41] --- Quantitative Factors and Construction Methods 1. Factor Name: Interaction of Factors and Predictors - **Factor Construction Idea**: Factor timing portfolios are constructed by interacting lagged predictors with factor returns, transforming the time-series prediction problem into a cross-sectional mean-variance optimization problem[17][19][23] - **Factor Construction Process**: 1. Define factor timing portfolios as: $$ G_{t}=X_{t-1}F_{t} $$ where $X_{t-1}$ are lagged predictors and $F_{t}$ are factor returns[19][20] 2. The expected return of the timing portfolio is: $$ E\left[G_{t}\right]=\mathrm{Cov}\left(X_{t-1},F_{t}\right) $$ indicating that timing returns depend on the predictive power of $X_{t-1}$ for $F_{t}$[21][22] 3. Construct $KJ$ timing portfolios for $K$ factors and $J$ predictors, always including the original factors as a baseline[22][23] - **Factor Evaluation**: The interaction approach effectively captures time-varying factor exposures, enabling robust timing strategies across diverse factor-predictor combinations[23][80] --- Factor Backtesting Results 1. Fama-French Factors (Standard Version) - **Mean Return**: 4.71% - **Standard Deviation**: 5.81% - **Sharpe Ratio**: 0.81 - **Appraisal Ratio**: 0.79 - **Worst 12-Month Return**: -5.62%[40][41] 2. Fama-French Factors (Large-Cap Version) - **Mean Return**: 3.65% - **Standard Deviation**: 6.49% - **Sharpe Ratio**: 0.56 - **Appraisal Ratio**: 0.54 - **Worst 12-Month Return**: -13.70%[40][41] 3. Jensen Factors (Small Predictor Set) - **Mean Return**: 2.97% - **Standard Deviation**: 2.01% - **Sharpe Ratio**: 1.48 - **Appraisal Ratio**: 1.51 - **Worst 12-Month Return**: -1.79%[40][41] 4. Jensen Factors (Large Predictor Set) - **Mean Return**: 2.73% - **Standard Deviation**: 1.91% - **Sharpe Ratio**: 1.43 - **Appraisal Ratio**: 1.46 - **Worst 12-Month Return**: -3.17%[40][41]

证券研究报告•深度/专题 影石创新投资探讨 ——25年前瞻专题之影石创新（688775.SH） 2025. 06. 11 华安证券研究所 分析师：邓 欣 S0010524010001 dengxin@hazq.com 分析师：成浅之 S0010524100003 chengqianzhi@hazq.com 联系人：唐楚彦 S0010124070002 tangchuyan@hazq.com 核心要点 影石创新坐拥三重产品壁垒全景优势+软件突出+迭代领先，以强产品力逐步抢占GoPro全球份额，最新销量已超越手持影 像设备昔日龙一GoPro；我们认为：从投资逻辑看，公司未来在产品外延、海外抢占、成本优化三方面皆有成长空间。 定价探讨：我们看好影石创新未来产品拓展空间与全球份额空间，我们测算25-26年公司收入78/101亿元（同比 +40%/+29%），利润13.4/16.5亿元（同比+35%/+23%），参考科技消费可比标的26年平均估值21X，公司当前价格存成 长性溢价空间（远期更存云增值、第二曲线等额外长期空间），关注公司长期价值。同步建议关注产业链标的：弘景光电、 韦尔股份、虹软科技、迅雷等（详见正文）。 ...

Group 1 - The overall situation of "fixed income +" funds shows a structured risk-return layering, with low-wave funds dominated by bonds (equity position <10%), providing stable but limited returns; medium-wave funds balanced between stocks and bonds (equity position 10%-20%), showing moderate volatility and robust performance in bull markets; high-wave funds with aggressive equity exposure (equity position >20%), exhibiting significant return elasticity but high drawdown risks [2][3][6] - The market is concentrated in mixed bond secondary funds, accounting for 45.33% of the total, reflecting investors' preference for a "steady progress" strategy, while bond-mixed funds attract high-risk capital through diversified asset allocation, although they represent a smaller scale [2][3][12] Group 2 - In the bond segment, low, medium, and high-wave "fixed income +" funds exhibit distinct risk-return characteristics: low-wave funds focus on high-rated, short-duration bonds, maintaining low drawdown days (<40 days) across market environments; medium-wave funds adopt a "coupon-focused, duration-adjusted" strategy, with drawdown days (40-80 days) and return volatility reflecting a "middle route" characteristic; high-wave funds employ aggressive duration management and credit downshifting, leading to significant drawdowns during bear markets [3][6][28] - The bond holdings of low-wave funds are primarily high-rated bonds, while medium-wave funds balance between high-rated credit bonds and 3-5 year interest rate bonds, and high-wave funds focus on cyclical and growth sectors, using bonds to control drawdowns while pursuing return elasticity [4][5][20] Group 3 - The stock allocation of low, medium, and high-wave "fixed income +" funds shows significant differences: low-wave funds maintain extremely low stock positions, with convertible bond allocations generally below 3%, focusing on high-rated credit bonds for stable coupon income; medium-wave funds allocate 10%-20% to stocks and convertible bonds, targeting stable sectors for excess returns; high-wave funds allow 20%-25% equity exposure, focusing on cyclical and growth sectors [4][5][6] - The performance of these funds during bull and bear markets reveals a clear differentiation in stock allocation strategies, with low-wave funds emphasizing stability, medium-wave funds focusing on dynamic balance, and high-wave funds pursuing return elasticity [6][20] Group 4 - The risk-return characteristics of the three types of funds vary significantly, with low-wave funds showing the least volatility and drawdown, medium-wave funds maintaining a balanced approach, and high-wave funds experiencing the highest volatility and drawdown during bear markets [6][22][23] - The analysis of bond holdings and duration strategies indicates that low-wave funds maintain a conservative duration profile, medium-wave funds adjust duration based on market conditions, and high-wave funds exhibit the most aggressive duration management [28][39][42]

Report Industry Investment Rating - The report maintains a neutral view on 50Y treasury bonds [7] Core View of the Report - In the current volatile market with increasing bullish bond market catalysts, the duration advantage of 50Y treasury bonds can be considered, but due to liquidity factors and changes in demand - side institutional behavior, and limited downward space for interest rates, a neutral view is maintained [7] Summary by Related Content Market Conditions and Opportunities for 50Y Treasury Bonds - Since June, trading opportunities in the bond market have gradually increased. Market reaction to Sino - US negotiations has dulled, May's fundamental data is likely to be favorable for the bond market, large banks have increased short - bond purchases, and the upcoming Lujiazui Forum may bring trading opportunities in the capital market [2] - Compared with other bonds, 50Y treasury bonds have a duration advantage in a bull market. Their supply and liquidity are weaker than 30Y bonds. The balance of treasury bonds with a remaining maturity of 45Y - 50Y is about 50.2 billion yuan, while that of 25Y - 30Y is over 2.2 trillion yuan. The current yield - to - maturity of 50Y treasury bonds is lower than that of 30Y local government bonds [2] - Another trading opportunity for 50Y treasury bonds is the interest rate elasticity after primary issuance. Since 2017, in nearly half of the 50Y treasury bond issuances, the primary issuance rate was higher than the secondary rate, often occurring in volatile or bear markets and related to the behavior of long - term bond - allocating institutions such as insurance companies [3][6] Demand - Side Analysis - Insurance institutions are the main buyers of ultra - long - term bonds. They have steadily increased their allocation of ultra - long - term bonds over the years, while funds may extend their duration at certain times for trading purposes [6] - June is a key time for insurance institutions to potentially increase their allocation of 50Y treasury bonds, but the allocation intensity this year may be lower than in previous years. The reasons include lower premium growth from January to April this year, high government bond issuance since the first quarter leading to more primary - market bond purchases by allocation funds, and insurance institutions' preference for 30Y local government bonds due to their higher coupon rates [6]

Group 1: Report Industry Investment Rating - Not mentioned in the provided content Group 2: Core Views of the Report - In May, the year-on-year CPI was -0.1%, with the decline remaining unchanged for three consecutive months, and the month-on-month CPI decreased by 0.2%, showing a seasonal decline. The year-on-year PPI was -3.3%, with the decline widening by 0.6 pct compared to April, and the month-on-month PPI decreased by 0.4%, with the decline remaining unchanged for three consecutive months [2]. - CPI is dragged down by the energy item, while core CPI continues to rise, indicating that demand may be warming up. The consumption structural improvement is still in the transition period, and service CPI continues to grow [3]. - International crude oil import factors, energy transformation, and insufficient demand for real estate and infrastructure lead to a decline in production material prices, dragging down the year-on-year negative growth of PPI to expand. However, PPI continues to show price improvement in high-tech industries, and the recovery impact gradually spreads from policy-driven areas to other consumer goods areas [3][4]. - From the perspective of residents' income, the year-on-year decline in rent has remained at -0.1% for three consecutive months since March, indicating that the income improvement trend may have stagnated. From the perspective of corporate activity signs, the year-on-year decline in pork prices for two consecutive months and the decline in liquor prices may indicate a decrease in corporate business activity [5][6]. - In the short term, CPI may face fluctuations, but in the long term, if external interference factors decrease marginally, CPI may break through upward. PPI is in a supply-side dilemma, but the effect of domestic demand pulling has initially appeared, and the "rush to export" of external demand may bring a phased rebound in industrial product prices [7][8]. Group 3: Summary by Relevant Catalogs Data Observation: Characteristics of May Inflation Data - CPI: Affected by the energy item, the year-on-year upward trend has stagnated below the critical line, but core inflation has risen to 0.6% year-on-year, showing that consumer demand continues to warm up. The consumption structure is improving, and service CPI has continued to grow for four months, reaching 0.5% year-on-year [2][3]. - PPI: Affected by international crude oil import factors, energy transformation, and insufficient demand for real estate and infrastructure, the year-on-year negative growth has expanded. However, PPI continues to show price improvement in high-tech industries, and the recovery impact gradually spreads from policy-driven areas to other consumer goods areas [3][4]. In - depth Perspective: Implications of March Price Data - Residents' income: The year-on-year decline in rent has remained at -0.1% for three consecutive months since March, indicating that the income improvement trend may have stagnated [5]. - Corporate activity: The year-on-year decline in pork prices for two consecutive months and the decline in liquor prices may indicate a decrease in corporate business activity [5][6]. Future Outlook: Trends Seen Through May Inflation Data - CPI: In the short term, it may face fluctuations due to external uncertainties, but in the long term, if external interference factors decrease marginally, it may break through upward [7]. - PPI: It is in a supply-side dilemma, but the effect of domestic demand pulling has initially appeared, and the "rush to export" of external demand may bring a phased rebound in industrial product prices, but attention should be paid to the negative impact of low - price competition among enterprises during the rush to export [8].

Investment Rating - Industry Investment Rating: Overweight [1] Core Views - The report highlights a significant increase in domestic large-scale energy storage tenders, indicating a growing market opportunity in the virtual power plant segment [1][5] - The photovoltaic sector is experiencing a decline in production in June, with intensified competition among various segments, suggesting a cautious outlook for the second quarter of 2025 [12][14] - The offshore wind sector is entering a favorable phase, with steady progress in offshore wind projects, particularly in Jiangsu and Fujian [20][21] - The hydrogen energy industry is developing positively, with an accelerated establishment of a supporting system for its growth [30][33] - The report emphasizes the importance of virtual power plants and new power system construction as key areas for investment [37][38] Summary by Sections Photovoltaics - Production in June is declining, leading to increased competition among different segments [12][14] - The overall price trend in the photovoltaic supply chain is stable, but there is downward pressure due to expected weak demand in the second half of the year [14][19] - Investment suggestions include focusing on companies with a strong position in the BC technology sector, such as Longi Green Energy and Aiko Solar [14][19] Wind Power - The offshore wind sector is seeing project approvals and progress, with significant projects in Jiangsu and Fujian [20][21] - Investment recommendations include focusing on undervalued stocks and those benefiting from offshore wind projects [23] Energy Storage - Domestic energy storage tenders exceeded 20 GWh in May, indicating a robust market [7][24] - The report suggests monitoring the recovery of large-scale energy storage market conditions [7][24] - The average price for two-hour energy storage systems reached a historical low of 0.55 yuan/Wh in May [28] Hydrogen Energy - The hydrogen energy sector is experiencing positive development, with a focus on transitioning from gray hydrogen to green hydrogen [30][33] - The report outlines plans for significant advancements in hydrogen energy technology and infrastructure by 2035 [33] Electric Grid Equipment - The report discusses the initiation of pilot projects for new power system construction, emphasizing the role of virtual power plants [37][38] - Investment opportunities are identified in companies involved in smart microgrids and virtual power plant technologies [37][38] Electric Vehicles - The potential cancellation of electric vehicle subsidies in the U.S. could negatively impact demand [39] - The report advises continued investment in high-profit companies within the electric vehicle supply chain [39] Humanoid Robots - The report notes significant developments in the humanoid robot sector, including leadership changes at Tesla's Optimus project and new product launches by various companies [42][46]