黄金、白银度过了不平凡的两个月。 先是白银逼空，逼着交易所提高保证金，限制开仓手数。紧接着迎来沃什时刻，白银大跌30%，黄金则遭遇1983年以来最大的单日跌幅。在这场14西格玛级 别的场景里，不少主观 CTA、宏观私募大幅回撤。 其中也包括各家"中国桥水"。火富牛数据显示，申九、金和晟、国恩、尚艺等私募的部分全天候产品，在2月初那周回撤7个点以上，澜音的全天候产品回撤 了20个点。 全天候策略配置了多个低相关性的资产，如果依据风险平价逻辑，低波动率的债券占比较高。在债券表现尚可的前提下，单周出现如此巨大的回撤，大概率 超配了金银。 | 股票类 | | 债券类 | | 商品类 | | | --- | --- | --- | --- | --- | --- | | 下证 50 | -0.93% | CFFEX2年期国债 | +0.05% | 南华商品指数 | -4.49% | | | | 期货 | | | | | 沪深 300 | -1.33% | CFFEX5年期国债 | +0.05% | 南华农产品指数 | -1.18% | | | | 期货 | | | | | 中证 500% | -2.68% | CFFE ...

报告导读： 本文将在宏观打分模型下，在战略投资组合的基础上捕捉资产短期内景气度变 化，构建以绝对收益为目标的大类资产 ETF 配置策略。 以下为本文所构建的部分大类资产ETF配置组合的历史净值与收益表现，组合的详细构建方法将在后续章节中展开阐述。 摘要 本篇报告的目的，是希望帮助绝对收益目标的机构进行资产配置，我们将构建具有稳定投资回报预期和历史高收益回撤比的资产配置策略，并希望能够获得预期年 化收益不小于 6% ，年化波动率不大于 5% ，最大回撤不大于 5% ，并且收益回撤比大于 1 的大类资产 ETF 配置组合。 本篇报告选取的大类资产，涉及 A 股权益、港股权益、债券、商品和境外权益 五类，各标的在公募基金市场中均具有对应且规模在同类中相对较大的 ETF 产品作 为可投资标的。我们基于选取的大类资产，构建 "境内股债资产" 和 "全球大类资产" 共两类资产配置模型，模型以每月末作为调仓时间，回测区间为 2017 年 1 月 1 日至 2026 年 2 月 13 日，并对单一资产和大类资产都设定了事前的权重上下限。 战略层：通过风险平价模型和 ES 风险平价模型设定资产权重中枢。 我们主要用风险平价模型 ...

ETF 配置系列（二）：宏观打分配置策略 [Table_Authors] 江涛(分析师) 以绝对收益为目标 多元配置为手段 本报告导读： 本文将在宏观打分模型下，在战略投资组合的基础上捕捉资产短期内景气度变化， 构建以绝对收益为目标的大类资产 ETF 配置策略。 投资要点： | | | 金 | | 021-23185672 | | --- | --- | | | jiangtao4@gtht.com | | 登记编号 | S0880525040067 | | | 倪韵婷(分析师) | | | 021-23185605 | | | niyunting@gtht.com | | 登记编号 | S0880525040097 | [Table_Report] 相关报告 高质量发展时代公募基金行业回顾与展望 2025.10.31 高质量发展时代公募基金行业回顾与展望 2025.10.29 恒生港股通科技主题指数投资价值分析 2025.10.15 "基"微成著系列（九） 2025.09.06 "基"海拾贝（四） 2025.08.05 证 券 研 究 报 告 请务必阅读正文之后的免责条款部分 基 金 评 价 基 专 题 报 ...

Group 1: Strategic Asset Pool Definition and Long-Term Characteristics - The report defines a global strategic asset allocation framework, covering equity, fixed income, alternative assets, and cash[1] - It analyzes long-term characteristics of equity assets in global, developed, and emerging markets, including sovereign and credit bonds, real estate, commodities, and private equity[1] - The analysis provides a solid data and theoretical foundation for subsequent return forecasts and portfolio construction[1] Group 2: Long-Term Economic Assumptions and Return Forecast Models - The report establishes long-term economic assumptions and return forecast models based on key macroeconomic variables such as economic growth, inflation, and interest rates[2] - It creates corresponding long-term return prediction models for various asset classes and estimates correlations and potential risk scenarios among different assets[2] Group 3: Strategic Portfolio Construction and Optimization - Strategic portfolio construction considers investor constraints and goal settings, including return targets, risk tolerance, liquidity needs, and regulatory/tax constraints[3] - Optimization methods include the classic mean-variance model, Black-Litterman model, Kelly-CVaR model, and risk parity model, with the mean-variance model and Kelly-CVaR showing superior long-term returns compared to single asset strategies[3] - The report emphasizes the importance of establishing a global market-weighted portfolio as a benchmark for strategic asset allocation[3] Group 4: Market Trends and Performance Metrics - The MSCI indices indicate that the U.S. market dominates with a weight of 64% in the MSCI ACWI index, followed by Japan at 4.9% and the UK at 3.3%[15] - The report highlights that the long-term volatility of MSCI EM is significantly higher than that of MSCI World, with both indices showing similar return patterns over time[23] - The Sharpe ratio for MSCI World and MSCI EM is similar, with long-term limits around -0.5 to +0.8, indicating comparable risk-adjusted returns[23]

Fixed Income Research - The net financing amount is at a historically high level, indicating that the logic of asset scarcity has dissipated. The overall change in the issuance guidance rates published by the trading association has mostly decreased by 5 to 1 basis points. In January, the issuance scale of credit bonds increased month-on-month, with only medium-term notes seeing a decrease in issuance amount, while other varieties saw increases. The net financing amount for credit bonds increased month-on-month, with medium-term notes showing a decrease, while other varieties saw increases. Corporate bonds, directional tools had negative net financing, while corporate bonds, medium-term notes, and short-term financing bonds had positive net financing [2][3]. - In the secondary market, the transaction scale of credit bonds decreased month-on-month, with transaction amounts for all varieties declining. The yield on credit bonds remained low and fluctuated, with most varieties showing a month-on-month decline in average yield. The credit spread for most varieties narrowed month-on-month, with the varieties that widened mainly concentrated in the 7-year term. Most varieties' spreads are at historical lows. From an absolute return perspective, insufficient supply and relatively strong allocation demand will continue to drive the recovery of credit bonds. Although fluctuations are inevitable due to various factors, the conditions for a comprehensive bear market in credit bonds remain insufficient. In the long run, future yields are still in a downward channel, and the strategy of increasing allocation during adjustments remains feasible [3]. Fund Research - In January, the market for actively managed equity funds saw a significant increase in issuance, with a total of 88 new funds issued, amounting to 91.48 billion yuan. The issuance of actively managed equity funds and passive equity funds was 41.70 billion units and 19.62 billion units, respectively, with a significant increase in the issuance of actively managed equity funds. Overall, the issuance market for equity funds has warmed up significantly, especially for actively managed equity funds [6][7]. - The performance of equity markets was outstanding in January, with all types of funds showing varying degrees of increase. The average increase for commodity funds was the largest at 17.92%. The growth style outperformed the value style, and the mid-cap balanced style had the largest increase at 8.99%, while the large-cap value style had the smallest increase at approximately 4.22% [8]. Industry Research - The valuation repair of the real estate chain can continue, with positive signals from the government regarding real estate policies. The market is transitioning from a large-scale expansion phase to a focus on quality improvement. The goal is to actively construct a new development model for real estate, emphasizing both short-term and long-term strategies. The sales recovery process will significantly impact bond valuations, and investors with a higher risk appetite may consider early positioning, especially in companies showing strong performance in new financing and sales recovery [4][10]. - In the paper industry, several leading companies have announced price increases for white cardboard and corrugated paper, with expected price hikes of 200 yuan/ton for white cardboard and 30-50 yuan/ton for corrugated paper. The upcoming annual maintenance period for paper companies will disrupt supply, while the approaching Spring Festival will boost packaging demand from e-commerce, food, and beverage sectors, supporting price increases [12]. - In the metals industry, the steel sector is expected to continue a weak performance due to the Spring Festival holiday, with production and demand both shrinking. The copper market is also anticipated to see inventory accumulation due to reduced production activities during the holiday, with a focus on post-holiday demand verification [13][15].

1. Report Industry Investment Rating - Not mentioned in the report 2. Core Viewpoints of the Report - In January, all major indices in the Shanghai and Shenzhen markets rose. The CSI 500 and ChiNext 50 led the gains, both rising by over 12%, while the SSE 50 had the smallest increase of 1.17%. Among the 31 Shenwan primary industries, 26 industries rose, with the top 5 gainers being non - ferrous metals, media, petroleum and petrochemicals, building materials, and chemicals. The declining industries were banking, household appliances, non - bank finance, transportation, and agriculture, forestry, animal husbandry, and fishery [1][14]. - In December 2025, the total number of new individual investor accounts in the market reached 2.5861 million, and the number of new institutional investor accounts was 11,100. The private securities investment fund market continued to heat up, with the newly registered scale in December increasing month - on - month to 54.174 billion yuan, and the existing scale reaching 22.15 trillion yuan [2][21][23]. - In January, 88 new funds were issued, with a total issuance scale of 91.481 billion yuan. The issuance shares of active equity funds and passive equity funds were 41.704 billion and 19.62 billion respectively, and the issuance shares of active equity funds increased significantly month - on - month. All types of funds rose to varying degrees, with commodity - type funds having the largest average increase of 17.92% [3][38]. - In January, the active equity funds increased their positions in the petroleum and petrochemical, non - ferrous metals, and basic chemical industries, and reduced their positions in the national defense and military industry, pharmaceutical biology, and computer industries. The overall position of active equity funds on January 30, 2025, was 73.88%, an increase of 0.78 percentage points from the previous month [4][54][58]. - In January, the ETF market had a net capital outflow of 841.187 billion yuan. Many ETFs related to the CSI 300 and SSE 50 indices in the broad - based index suffered significant capital outflows. Among the actively traded individual securities, gold stock ETFs, China - South Korea semiconductor ETFs, mining ETFs, industrial non - ferrous metal ETFs, and ChiNext chip design ETFs led the gains, rising by 22.5% - 39.6%, while bank ETFs, E Fund's Hong Kong Stock Connect medical ETF, automobile ETFs, Peng Hua's general aviation ETF, and Huaxia's financial real - estate ETF led the losses, falling by 4.8% - 6.4% [5][61][62]. - In January, the risk - parity model rose by 2.07%, and the risk - budget model rose by 2.39% [6][73]. 3. Summary According to Relevant Catalogs 3.1 Domestic Market Situation - In January, all major stock indices in the Shanghai and Shenzhen markets rose. The CSI 500 and ChiNext 50 led the gains, both rising by over 12%, and the SSE 50 had the smallest increase of 1.17%. Among the 31 Shenwan primary industries, 26 industries rose, and 5 industries fell. In the bond market, the ChinaBond Composite Total Return Index rose by 0.22%, and the ChinaBond Treasury, financial bond, and credit bond total return indices fluctuated between a decline of 0.09% and an increase of 0.24%. The CSI Convertible Bond Index rose by 5.82%. In the commodity market, the Nanhua Commodity Index rose by 8.61% [1][14]. - In December 2025, the total number of new individual investor accounts in the market reached 2.5861 million, and the number of new institutional investor accounts was 11,100. The private securities investment fund market continued to heat up, with the newly registered scale in December increasing month - on - month to 54.174 billion yuan, and the existing scale reaching 22.15 trillion yuan [2][21][23]. 3.2 European, American, and Asia - Pacific Market Situation - In January, most major indices in the European, American, and Asia - Pacific markets rose. In the US stock market, the S&P 500 rose by 0.29%, the Dow Jones Industrial Average rose by 1.76%, and the Nasdaq Composite rose by 0.95%. In the European market, the French CAC40 fell by 0.28%, and the German DAX rose by 0.20%. In the Asia - Pacific market, the Hang Seng Index rose by 6.85%, and the Nikkei 225 rose by 5.93% [26]. 3.3 Market Valuation Situation - In January, the valuations of all major market indices rose. In terms of the historical quantile of price - to - earnings ratio, the CSI All - Share Index led the increase, rising by 8.5 percentage points. In terms of the historical quantile of price - to - book ratio, the CSI 1000 led the increase, rising by 15.3 percentage points. Among the industries, the top five industries with the highest historical quantile of price - to - earnings ratio of the Shenwan primary index last month were real estate, electronics, chemicals, commercial trade, and comprehensive. The price - to - earnings ratio quantile of the real estate industry was at a high level, and that of the electronics industry reached 96.2%. The bottom five industries with the lowest historical quantile of price - to - earnings ratio were non - bank finance, agriculture, forestry, animal husbandry, and fishery, food and beverages, beauty care, and pharmaceutical biology. The valuation of the non - bank finance industry was close to its historical low since 2013 [31]. 3.4 Overall Situation of Public - Offering Funds 3.4.1 Fund Issuance Situation - In January, 88 new funds were issued, with a total issuance scale of 91.481 billion yuan. Among them, 36 stock - type funds were issued with a scale of 19.669 billion yuan; 36 hybrid funds were issued with a scale of 46.646 billion yuan; 4 bond - type funds were issued with a scale of 49.46 billion yuan; 10 FOF funds were issued with a scale of 18.713 billion yuan; and 2 QDII funds were issued with a scale of 15.07 billion yuan. The issuance shares of active equity funds and passive equity funds were 41.704 billion and 19.62 billion respectively, and the issuance shares of active equity funds increased significantly month - on - month [38]. 3.4.2 Fund Market Return Situation - In January, the equity market performed outstandingly, and all types of funds rose to varying degrees. Commodity - type funds had the largest average increase of 17.92%. From the perspective of fund style indices, in January, the market showed a general upward trend, and the market performance of different - style funds was differentiated. The growth style outperformed the value style, and the small - and medium - cap style outperformed the large - cap style. Overall, the mid - cap balanced style had the largest increase of 8.99%, and the large - cap value style had the smallest increase of about 4.22%. Among different - scale equity - biased public - offering funds, the mini - funds with a scale of 50 million - 100 million had the largest average increase of 7.27% and a positive return ratio of 95.25%, while the super - large funds with a scale of over 10 billion had the smallest average increase of 4.79% and a positive return ratio of 87.10% [3][42][51]. 3.4.3 Active Equity Fund Position Situation - In January, active equity funds increased their positions in the petroleum and petrochemical, non - ferrous metals, and basic chemical industries, and reduced their positions in the national defense and military industry, pharmaceutical biology, and computer industries. The overall position of active equity funds on January 30, 2025, was 73.88%, an increase of 0.78 percentage points from the previous month [4][54][58]. 3.5 ETF Fund Situation - In January, the ETF market had a net capital outflow of 841.187 billion yuan. Among them, stock - type ETFs had a net outflow of 793.799 billion yuan, cross - border ETFs had a net inflow of 30.917 billion yuan, and bond - type ETFs had a net outflow of 106.172 billion yuan. In terms of liquidity, the average daily trading volume of the overall ETF market in this period reached 604.575 billion yuan, the average daily trading volume reached 226.327 billion shares, and the average daily turnover rate was 9.17%, an increase of 2.02 percentage points from December of the previous year. Many ETFs related to the CSI 300 and SSE 50 indices in the broad - based index suffered significant capital outflows. Among the actively traded individual securities, gold stock ETFs, China - South Korea semiconductor ETFs, mining ETFs, industrial non - ferrous metal ETFs, and ChiNext chip design ETFs led the gains, rising by 22.5% - 39.6%, while bank ETFs, E Fund's Hong Kong Stock Connect medical ETF, automobile ETFs, Peng Hua's general aviation ETF, and Huaxia's financial real - estate ETF led the losses, falling by 4.8% - 6.4%. The ETFs with the largest net capital inflows were non - ferrous metal ETFs, gold ETFs, chemical ETFs, power grid equipment ETFs, and semiconductor equipment ETFs, while the ETFs with the largest net capital outflows were Huatai - Peregrine CSI 300 ETF, E Fund CSI 300 ETF, Huaxia CSI 300 ETF, SSE 50 ETF, and Harvest CSI 300 ETF [5][61][62]. 3.6 Model Operation Situation - Four types of large - scale asset allocation models were constructed using stocks, bonds, commodities, and QDII. In January, the risk - parity model rose by 2.07%, and the risk - budget model rose by 2.39%. Since 2015, the annualized return of the risk - parity model has been 4.94%, with a maximum drawdown of 2.31%, and the annualized return of the risk - budget model has been 5.13%, with a maximum drawdown of 9.80%. Next month, the asset allocation weights of the models remain unchanged. For the risk - parity model, stocks: bonds: commodities: QDII = 6%: 69%: 12%: 13%; for the risk - budget model, stocks: bonds: commodities: QDII = 14%: 49%: 8%: 29% [68][73][74].

1. Report Industry Investment Rating The provided content does not include the industry investment rating. 2. Core View of the Report The report focuses on the localization dilemma of the traditional risk parity model with "low volatility and low returns" in a low - interest - rate environment. By introducing the "Relaxed Risk Parity (RRP)" framework and a dynamic return anchoring mechanism, it constructs an all - weather enhanced strategy that does not rely on macro - timing and balances "risk diversification" and "return elasticity", achieving a logical advancement from localization adaptation to global allocation [1]. 3. Summary According to Relevant Catalogs 3.1 Introduction - In the low - interest - rate era, the traditional single - asset investment framework faces challenges. Asset allocation has become a core source of returns. The mainstream asset - allocation models include the mathematical optimization system and the macro - cycle system, but they face "acclimatization" in the Chinese market. The RRP framework is introduced to solve the problems [15][16][18]. 3.2 Global - scale Main Asset - Allocation Model Introduction - **Logic Divergence and Evolution Path of Mainstream Models**: The mainstream models are divided into the mathematical optimization system starting from the Markowitz mean - variance model and the macro - cycle system represented by the Merrill Lynch Clock. The former evolves from capital allocation to risk allocation, and the latter deepens from state segmentation to timing rotation [20][21][22]. - **China - adaptation Process of Asset - Allocation Models**: The traditional Merrill Lynch Clock has "acclimatization" in the Chinese market. A currency - credit model is proposed as a local alternative. The risk - parity strategy also has a localization dilemma due to the lack of leverage tools and hedging products, and the RRP framework is a solution [26][30]. 3.3 What is Risk Parity? - **Allocation and Hedging, Risk Parity and Factor Parity**: Risk parity is an asset - allocation philosophy based on "risk budgeting". It has evolved into the asset/macro parity mode represented by Bridgewater and the factor parity mode represented by AQR and academia [32]. - **Logical Architecture and Core Calculation Rules**: Based on Euler's theorem, the risk - parity strategy decomposes the total portfolio risk into the risk contributions of each asset and makes them equal through optimization algorithms [33]. 3.4 Relaxed Improvement of Risk Parity - **Logical Reasoning and Core Modeling of RRP**: The traditional risk - parity model has defects such as over - reliance on low - volatility assets and loss of mean - variance efficiency. The RRP framework relaxes the hard constraint of equal risk contributions to a soft - penalty term and constructs a comprehensive optimization model [43][44]. - **Back - test Results**: - **From "Defensive Trap" to "Efficiency Leap"**: Compared with the standard risk - parity portfolio (V1), the relaxed risk - parity portfolio (V2) has significantly improved in terms of annualized return, Sharpe ratio, winning rate, and has more flexible bond - leverage use and dynamic portfolio structure [70]. - **From "Local Trial - and - Error" to "Global Allocation"**: The globalized relaxed risk - parity portfolio (V3) has a diversified structure and lower trading friction. It can use the dislocation of Sino - US economic cycles to achieve better performance [76][79]. - **Generalization Ability of the Model under Different Parameter Windows**: When the parameter - estimation window is extended from 1 year to 2 years, the RRP model still shows better risk - return characteristics, proving its long - term effectiveness [85]. 3.5 Subsequent Strategy Optimization - **AI - enabled**: Introduce a deep - reinforcement learning framework to construct an intelligent agent that can dynamically adjust core parameters according to the macro - environment [86]. - **Response to Extreme Environments**: Use the idea of volatility parity and graph - theory algorithms to improve the robustness of the strategy in extreme environments [88]. - **Paradigm Reconstruction**: Shift from asset - based to factor - based risk control, and construct a three - dimensional allocation system [89].

Investment Rating - The report indicates a positive investment rating for the industry, suggesting an "Overweight" position relative to the CSI 300 index, with expected returns exceeding 15% [36]. Core Insights - The report highlights that gold has once again led the gains among major asset classes, with global asset allocation models recording positive returns. The domestic asset BL models showed returns of 0.28% and 0.26%, while global models recorded returns of 0.14% and 0.12% for the week [1][2][4]. Summary by Sections 1. Major Asset Performance Tracking - For the week of January 12 to January 16, 2026, major asset performances were as follows: SHFE gold increased by 2.57%, Hang Seng Index by 2.23%, and CSI 1000 by 1.27%. Conversely, the CSI 300 and S&P 500 saw declines of 0.57% and 0.45% respectively [7][10]. 2. Major Asset Allocation Strategy Tracking - The report details the performance of various quantitative asset allocation models. The domestic asset BL model 1 achieved a weekly return of 0.26%, while model 2 achieved 0.28%. The global asset BL model 1 and 2 recorded returns of 0.12% and 0.14% respectively for the same week [10][17][21]. 2.1. BL Model Strategy Tracking - The domestic asset BL model 1 has a year-to-date return of 1.13% with an annualized volatility of 2.85%. The global asset BL model 1 has a year-to-date return of 0.69% with an annualized volatility of 2.9% [17][18]. 2.2. Risk Parity Model Strategy Tracking - The domestic risk parity model reported a weekly return of 0.20% and a year-to-date return of 0.49%, with an annualized volatility of 1.16%. The global risk parity model achieved a weekly return of 0.13% and a year-to-date return of 0.38% [21][22]. 2.3. Macro-Factor Based Asset Allocation Strategy - The macro-factor based asset allocation strategy yielded a weekly return of 0.23% and a year-to-date return of 0.61%, with an annualized volatility of 1.73% [29].

Group 1 - The core point of the article highlights the impressive performance of Bridgewater Associates, the world's largest hedge fund, which achieved record returns in 2025, breaking a trend of annual returns below 3% from 2012 to 2024 [1][18] - Bridgewater's All Weather strategy focuses on risk parity, balancing risk contributions rather than simply allocating funds evenly across different assets, which has led to strong performance in various market conditions [19][21] - Domestic alternatives to Bridgewater's strategy have emerged, with many private equity firms adopting risk parity models and achieving returns between 20% to 40% in 2025 [19][20] Group 2 - The China Europe Wealth Management's Multi-Asset All Weather strategy achieved a return of 10.78% over the past year, significantly outperforming its benchmark of 5.72%, with a maximum drawdown of only 1.94% and a Sharpe ratio of 3.77 [22] - The service offerings for high-net-worth clients include a comprehensive suite of products, featuring over 80 private equity products from more than 30 top managers, covering various mainstream strategies [25][26] - A dedicated remote advisory team of over 30 members, with an average of more than 8 years of experience, has provided personalized investment solutions and ongoing support to thousands of high-net-worth clients, achieving a customer satisfaction rate exceeding 90% [28][30] Group 3 - The investment strategy includes continuous tracking and personalized asset diagnosis reports, which help clients navigate market fluctuations and avoid irrational decisions during periods of volatility [30][33] - Recent events, such as the "Insight 2026" investment strategy conference, have facilitated direct communication between investors and fund managers, enhancing the understanding of market dynamics and investment strategies [32][33] - The overall service model aims to provide clients with a comprehensive investment experience, ensuring they are well-informed and supported throughout their investment journey [33]

Report Industry Investment Rating - No investment rating provided in the report. Core Viewpoints - In 2026, the scale of macro - hedge strategies is expected to increase further as their allocation value is increasingly recognized in the market. Risk - parity strategies will play a stronger role as the base position in the portfolio, and the returns of risk - parity managers will experience a certain degree of mean reversion. [36][37] - The performance of subjective CTA strategies in 2026 will be better than that in 2025. The decrease in Sino - US macro uncertainties and the increase in commodity volatility in a low - interest - rate environment will benefit subjective CTA managers. [58] Summary by Directory 01 Macro - Hedge Strategy Research and Outlook Manager Classification and Characteristics - Macro - hedge managers are classified into three types: risk - parity, asset - rotation, and multi - asset multi - strategy. This report focuses on the first two types. Risk - parity managers use the risk - parity model as the basis and enhance it, with relatively consistent performance; asset - rotation managers are based on asset - rotation frameworks like the Merrill Lynch Clock, emphasizing asset timing allocation and having less consistent performance. [6] Domestic Manager Performance in 2025 - As of November 28, 2025, the net value of the "risk - parity" macro - hedge index was 1.172, and that of the "asset - rotation" index was 1.101. In the 46 weeks from January 3 to November 28, 2025, risk - parity managers had positive weekly returns in 30 weeks and negative returns in 16 weeks, with the largest single - week drawdown occurring after the Tomb - Sweeping Festival on April 11. Asset - rotation managers had positive weekly returns in 25 weeks and negative returns in 20 weeks, with the largest single - week drawdown occurring in the week of November 21. In the context of global supply - chain reshaping, risk - parity managers outperformed asset - rotation managers in 2025. [10] Asset Correlation Analysis - In 2025, the negative correlation between treasury bonds and equity indices weakened compared to the end of last year. The China Securities Commodity Index was positively correlated with stock indices and negatively correlated with treasury bonds and gold. Gold, as a safe - haven asset, had a stronger correlation with treasury bonds. There were significant differences in the performance correlations of risk - parity and asset - rotation managers with equity, treasury bonds, and gold. [13] - In terms of equity assets, the correlation between the risk - parity strategy and the CSI 300 was 0.230, and that with the CSI 1000 was 0.186. The correlations of the asset - rotation strategy with the CSI 300 and CSI 1000 were 0.628 and 0.641 respectively. The asset - rotation strategy's returns were more dependent on stocks, and the large drawdown in the week of November 21 was related to the stock decline. [19] - After a five - fold leverage treatment of 10 - year treasury bonds, the correlation between the risk - parity strategy and 10 - year treasury bond futures was 0.221, while that of the asset - rotation strategy was - 0.068. Many managers believed that the treasury bond market was in a bear market, so asset - rotation managers mostly reduced or shorted treasury bonds, while risk - parity managers still held bond positions. [23] - In 2025, gold was one of the strongest - performing assets, with a cumulative net value of the Gold ETF of 1.588 from January 3 to November 28. The correlation between the risk - parity strategy and the Gold ETF was 0.453, while that of the asset - rotation strategy was 0.110. Gold had a greater impact on risk - parity strategies. [26] Overseas Manager Performance in 2025 - As of October 2025, the net value of the "unidentified" macro - hedge index was 1.088, the "subjective" macro - hedge index was 1.129, and the "quantitative" macro - hedge index was 1.159. Quantitative macro - hedge strategies performed the best, followed by subjective strategies, similar to the domestic situation. The maximum drawdowns of the unidentified and quantitative macro - hedge strategies occurred in April, indicating that domestic risk - parity managers may use similar underlying models to overseas ones. [29] - The unidentified macro - hedge strategy index had a more balanced correlation with various asset classes, with a near - zero correlation with New York gold. The subjective macro - hedge index had a high correlation of 0.792 with the S&P 500 and a negative correlation with New York gold, indicating that its returns were more dependent on the US stock market. The quantitative macro - hedge strategy also had a high correlation of 0.627 with the S&P 500 and a relatively high correlation of 0.300 with the S&P GSCI, but a negative correlation with US treasury bonds and gold. [33] Outlook for 2026 - The scale of macro - hedge strategies will increase as their allocation value is recognized. Some investors may replace part of their stock - neutral strategy allocation with low - volatility macro - hedge strategies. The role of risk - parity strategies as the base position in the portfolio will be enhanced, and their return attribution is relatively clear. [36] - The returns of risk - parity managers will experience mean reversion in 2026. Since the probability of bonds and gold replicating their price increases since 2024 is significantly reduced, the returns of these managers will decline. Historically, the long - term return of the basic risk - parity model is around 6 - 8%. [37] 02 Discretionary CTA Strategy Research and Outlook Performance in 2025 - The net value performance of managers in the observation pool in 2025 was weaker than in the same period of 2024. Uncertainties in Sino - US trade friction reduced the trading certainty of discretionary CTA managers based on industrial supply - demand research, weakening their position - holding confidence and return - generating ability. After June, although market sentiment improved, the lack of improvement in the industrial sector led to significant drawdowns for many managers, lowering the annual return. [40] Sector - Specific Performance - Black - sector managers showed some resilience in returns in 2025. In the first half of the year, the collapse of coal costs led to a downward trend in the black - sector prices, with good persistence and low volatility. The concerns about external demand due to Sino - US trade friction coincided with the seasonal decline in coal prices, providing trading opportunities with industrial and macro resonance. In the second half of the year, differences in the implementation of anti - involution policies led to a negative view among industrial - based managers, resulting in significant drawdowns. [45] - Agricultural - product managers were greatly affected by trade frictions between China and the US, Canada, etc. The unpredictable changes in agricultural - product imports and price fluctuations made it difficult for them to generate returns. [45] Industry Changes - Leading managers are iterating towards multi - asset and multi - strategy models. The limited capital capacity of single - asset futures trading, the need to understand the trading behavior of other market participants, and the benefits of multi - asset diversification are the main reasons. [50] - Start - up private - equity funds have shown strong drawdown - control ability since their establishment. Compared with the past, current start - up discretionary CTA private - equity funds have a clearer understanding of investors' risk preferences and a more explicit performance - oriented approach, enabling them to enter institutional investors' asset - allocation pools more quickly. [52] - In a diversified market structure, single - industry logic is insufficient for trading. Managers need to have comprehensive capabilities in macro - judgment, trading, and risk - control. Research determines the winning rate, trading and risk - control determine the profit - loss ratio, and an excellent trader may not be an excellent asset - management manager. [55] Outlook for 2026 - The performance of discretionary CTA strategies in 2026 will be better than in 2025. The decrease in Sino - US macro uncertainties will make commodity supply - demand the dominant factor in trading, and the increase in commodity volatility in a low - interest - rate environment will be beneficial for managers to generate returns. The increase in the scale of discretionary CTA managers based on industrial research will also contribute to the strength of industrial logic in the market. [58]