证券研究报告 | 债券市场专题研究 | 债券研究 债券市场专题研究 报告日期：2025 年 08 月 22 日 基于卡玛比率的低收益高波动下债市应对策略 ——债市策略思考 核心观点 债市仍偏逆风，但低收益高波动的另一面是"错误定价"频现，建议投资者"低仓位 +高胜率"做好防守反攻，快进快出博弈超跌反弹机会。 ❑ 2025 年以来债市处于低收益、高波动状态 不同于过去一年债市的高收益、高波动状态，2025 年以来债市已进入低收益、高 波动状态：一方面，2025 年以来 10 年国债收益率波动明显加大；另一方面，从 基金业绩表现来看，2025 年以来基金回报率较往年有所下降。 ❑ 如何评价基金投资的性价比？ 分析师：郑莎 执业证书号：S1230524080012 zhengsha@stocke.com.cn 相关报告 1 《赎回潮行情何时至右侧？》 2025.08.20 2 《探析"反内卷"行情对利率 中枢的影响》 2025.08.18 3 《转债延续上涨，关注产业趋 势》 2025.08.17 （1）卡玛比率比夏普比率更能反映真实风险，但二者需结合使用。夏普比率更适 用于中短债及纯债类波动小，回撤较低的基金，而 ...

1. Report Industry Investment Rating No relevant content provided. 2. Core Viewpoints of the Report The report focuses on the performance comparison benchmarks of domestic active bond funds. It analyzes the benchmark settings of various sub - types of active bond funds and their deviations in actual operations. After the release of the "Action Plan", some bond funds have adjusted their performance comparison benchmarks. The report also studies the correlation between funds and benchmarks, tracking errors, and excess returns [1][9]. 3. Summary According to the Table of Contents 3.1 Introduction On May 7, 2025, the CSRC issued the "Action Plan for Promoting the High - quality Development of Public Funds", emphasizing the importance of performance comparison benchmarks. The report, as the second in the series, will analyze the benchmark settings and actual operation deviations of domestic active bond funds [9]. 3.2 Active Bond Fund Performance Comparison Benchmark Characteristics - **Generalized Active Bond Fund Sample Selection**: As of May 7, 2025, 4191 generalized active bond funds in existence and with performance comparison benchmarks were selected as samples, with a total scale of 9.05 trillion yuan. The samples include 7 types of funds, and the medium - long - term pure - bond funds have the largest number and scale [9][10][12]. - **Performance Benchmark Composition Method**: The performance comparison benchmarks of active bond funds have various forms, mainly including single bond indexes or weighted composites of different indexes. The component indexes can be classified into 6 major categories, and the bond index can be further divided into 5 sub - types, while the stock index can be divided into 9 sub - types [13]. - **Performance Benchmark Commonly Used Indexes**: The top ten "main benchmark indexes" of medium - long - term pure - bond funds are mainly indexes compiled by ChinaBond. For example, the number of funds with ChinaBond - Composite Full Price (Total Value) Index as the main benchmark index is 964, accounting for 52.56%. The main benchmark indexes of convertible bond funds are mainly convertible bond indexes, with CSI Convertible Bond Index being the most used. The main benchmark indexes of fixed - income enhanced funds are mainly A - share market indexes such as CSI 300 Index [36][41][50]. - **Comparison of Commonly Used Index Clusters**: The ChinaBond index system is compiled by the Central Government Bond Depository Trust & Clearing Co., Ltd., and the CSI index system is compiled by CSI Index Co., Ltd. The component bond listing locations, remaining maturities, and credit ratings of ChinaBond and CSI indexes are different [54][57]. - **Weight Distribution of "Main Benchmark Indexes"**: For most active bond funds, the weights of ChinaBond - Composite Full Price (Total Value) Index and ChinaBond - Composite Wealth (Total Value) Index are mainly in the range of 90 - 100% for medium - long - term pure - bond funds, mixed bond - type first - level funds, and mixed bond - type second - level funds. The weights of equity indexes in the performance comparison benchmarks of mixed bond - type second - level funds, convertible bond - type funds, and partial - debt hybrid funds are relatively concentrated [62][66]. 3.3 Fund Performance and Benchmark Correlation and Other Analyses - **Correlation Analysis between Active Bond Funds and Their Benchmarks**: From 2022 to 2025, convertible bond - type funds, short - term pure - bond funds, medium - short - term pure - bond funds, medium - long - term pure - bond funds, and partial - debt hybrid funds have relatively high correlations with their performance comparison benchmarks, while mixed bond - type first - level funds and mixed bond - type second - level funds have relatively low correlations [72][73]. - **Tracking Error and Excess Return of Funds Relative to the Benchmark**: The average tracking error of pure - bond funds is less than that of products with embedded options. Among fixed - income enhanced bond funds, first - level bond funds have lower tracking errors, second - level bond funds and partial - debt hybrid funds are relatively close, and convertible bond funds have the highest and most volatile tracking errors. Most pure - bond funds can outperform the benchmark in most years, and the average outperformance is within 2%. Among fixed - income enhanced funds, partial - debt hybrid funds have relatively high average excess returns [3][78]. - **Distribution of Fund Types with Significant Underperformance against the Benchmark**: Pure - bond funds have relatively small deviations from the benchmark and a low proportion of significant underperformance. The performance of fixed - income enhanced funds is related to the selected time interval and the performance of the equity market. In the long - term, active bond funds have the ability to obtain positive excess returns relative to the benchmark, but there are significant performance differences within each type of fund [3].