大类资产及择时观点月报（2025.10） [Table_Authors] 郑雅斌(分析师) 债市观点发生改变 本报告导读： 根据 2025 年 9 月底的最新数据，股票、债券和黄金市场在 2025 年 10 月信号分别为 正向，负向和正向。 投资要点： 金融工程 /[Table_Date] 2025.10.09 | | 021-23219395 | | --- | --- | | | zhengyabin@gtht.com | | 登记编号 | S0880525040105 | | | 曹君豪(分析师) | | | 021-23185657 | | | caojunhao@gtht.com | | 登记编号 | S0880525040094 | [Table_Report] 相关报告 大类资产及择时观点月报（2025.09） 2025.09.01 风格及行业观点月报（2025.09） 2025.09.01 大类资产及择时观点月报（2025.08） 2025.08.01 风格及行业观点月报（2025.08） 2025.08.01 大类资产及择时观点月报（2025.07） 2025.07.02 证 券 研 究 报 ...

Quantitative Models and Construction Methods 1. Model Name: "Clock + Turning Point Improvement Method" Asset Rotation Model - **Model Construction Idea**: This model integrates macroeconomic risk factors with asset rotation strategies, leveraging the "investment clock" concept and improving turning point identification through a combination of momentum and phase judgment methods [8][23][24] - **Model Construction Process**: 1. Macro risk factors (e.g., economic growth, inflation, interest rates, credit, exchange rates, and term spreads) are used to determine the macroeconomic state [8] 2. The "investment clock" framework is applied to link macroeconomic states with asset performance. For example, recovery and overheating phases favor equities and commodities, while stagflation and recession phases favor bonds and gold [9][15] 3. Turning points in macroeconomic factors are identified using a combination of momentum and phase judgment methods: - Momentum is calculated as: $$ Momentum_t = X_t - \frac{1}{3}(X_{t-1} + X_{t-2} + X_{t-3}) $$ where \( X_t \) represents the macro factor value at time \( t \) [16] - Phase judgment uses a 38-month sine wave to identify the current phase of macro factors, categorizing them into upward, downward, top, or bottom regions [21][22] 4. Asset scores are calculated based on the macro factor states, and risk allocation is adjusted accordingly. Initial risk weights are set as large-cap stocks: small-cap stocks: bonds: commodities: gold = 1:1:1:0.5:0.5. Positive scores double the risk allocation, while negative scores halve it [24] 5. Backtesting is conducted over the period from January 2011 to December 2023 [25] - **Model Evaluation**: The model demonstrates strong performance in terms of returns, risk control, and drawdown management, achieving nearly 10% annualized returns while maintaining low volatility and drawdowns [27] --- Model Backtesting Results 1. "Clock + Turning Point Improvement Method" Asset Rotation Model - **Total Return**: 242.45% - **Annualized Return**: 9.93% - **Annualized Volatility**: 6.83% - **Sharpe Ratio**: 1.45 - **Maximum Drawdown**: 6.31% - **Win Rate**: 73.08% [27] 2. Benchmark Equal-Weighted Portfolio - **Total Return**: 83.59% - **Annualized Return**: 4.78% - **Annualized Volatility**: 10.99% - **Sharpe Ratio**: 0.43 - **Maximum Drawdown**: 20.63% - **Win Rate**: 55.77% [27] --- Quantitative Factors and Construction Methods 1. Factor Name: Macro Risk Factors - **Factor Construction Idea**: These factors aim to capture various dimensions of macroeconomic risks, including economic growth, inflation, interest rates, credit, exchange rates, and term spreads, providing a comprehensive view of the macroeconomic environment [8] - **Factor Construction Process**: 1. **Economic Growth**: - Indicators: Industrial production YoY, PMI, retail sales YoY - Processing: HP filtering and volatility-weighted averaging - Interpretation: Positive values indicate economic expansion [8] 2. **Inflation**: - Indicators: PPI YoY, CPI YoY - Processing: HP filtering and volatility-weighted averaging - Interpretation: Positive values indicate rising inflation [8] 3. **Interest Rates**: - Indicators: Bond indices (1-3 years), money market indices - Processing: Equal-weighted portfolio construction and net value calculation - Interpretation: Negative values indicate falling interest rates and loose monetary conditions [8] 4. **Exchange Rates**: - Indicators: Gold prices (Shanghai and London) - Processing: Equal-weighted long-short portfolio construction - Interpretation: Positive values indicate currency depreciation [8] 5. **Credit**: - Indicators: Corporate bond indices (AAA) vs. government bond indices - Processing: Duration-neutral portfolio construction - Interpretation: Positive values indicate widening credit spreads and tighter credit conditions [8] 6. **Term Spreads**: - Indicators: Short-term vs. long-term bond indices - Processing: Duration-neutral portfolio construction - Interpretation: Positive values indicate widening term spreads [8] --- Factor Backtesting Results 1. Macro Risk Factors (July 2025 State) - **Economic Growth**: Upward (Recovery phase) - **Inflation**: Downward - **Interest Rates**: Downward - **Credit**: Downward - **Exchange Rates**: Downward - **Term Spreads**: Downward [36]

Quantitative Models and Construction Methods Model Name: Macro Risk Factor-Based Asset Rotation Model - **Model Construction Idea**: The model utilizes macroeconomic risk factors to guide asset allocation decisions, aiming to optimize returns while controlling risks[5][8]. - **Model Construction Process**: 1. **Macro Risk Factor System**: Six macro risk factors are constructed using economic growth, inflation, interest rates, exchange rates, credit, and term spread indicators[8]. 2. **Investment Clock**: The model incorporates the "growth-inflation clock" and "interest rate-credit clock" to understand asset performance under different macroeconomic conditions[9][10][11]. 3. **Phase Judgment Method**: The model uses factor momentum and phase judgment methods to identify macroeconomic turning points[16][17][21][22]. 4. **Asset Rotation Model**: Combining the investment clock and phase judgment methods, the model adjusts asset allocation based on current macroeconomic conditions[23][24]. 5. **Backtesting Period**: The model is backtested from January 2011 to December 2023[25]. 6. **Performance Metrics**: The model's performance is evaluated using total return, annualized return, annualized volatility, Sharpe ratio, maximum drawdown, and win rate[27]. - **Model Evaluation**: The model demonstrates excellent performance in terms of returns, risk control, and drawdown management, achieving nearly 10% annualized returns with controlled risk exposure[27]. Model Backtesting Results - **Macro Risk Factor-Based Asset Rotation Model**: - Total Return: 242.45%[27] - Annualized Return: 9.93%[27] - Annualized Volatility: 6.83%[27] - Sharpe Ratio: 1.45[27] - Maximum Drawdown: 6.31%[27] - Win Rate: 73.08%[27] Quantitative Factors and Construction Methods Factor Name: Macro Risk Factors - **Factor Construction Idea**: The factors are designed to capture various aspects of the macroeconomic environment, providing a comprehensive risk perspective[8]. - **Factor Construction Process**: 1. **Economic Growth**: Constructed using industrial added value, PMI, and retail sales growth, processed with HP filtering and weighted by volatility inverse[8]. 2. **Inflation**: Constructed using PPI and CPI growth, processed with HP filtering and weighted by volatility inverse[8]. 3. **Interest Rates**: Constructed using bond indices and money market fund indices, weighted equally[8]. 4. **Exchange Rates**: Constructed using gold prices in Shanghai and London, forming an equal-weighted long-short portfolio[8]. 5. **Credit**: Constructed using corporate bond and government bond indices, forming a duration-neutral portfolio[8]. 6. **Term Spread**: Constructed using short-term and long-term bond indices, forming a duration-neutral portfolio[8]. - **Factor Evaluation**: The factors provide a detailed and comprehensive view of macroeconomic risks, aiding in better asset allocation decisions[8]. Factor Backtesting Results - **Economic Growth Factor**: Upward trend[36] - **Inflation Factor**: Downward trend[36] - **Interest Rate Factor**: Tight interest rate, loose credit[36] - **Credit Factor**: Downward trend[36] - **Exchange Rate Factor**: Downward trend[36] - **Term Spread Factor**: Downward trend[36] Monthly Performance Review (May 2025) - **Model Performance**: - Monthly Return: -0.29%[30] - Benchmark Return: 0.3%[30] - Excess Return: -0.6%[30] - **Asset Allocation**: - Large Cap Stocks: 5.3%[34] - Small Cap Stocks: 3.01%[34] - Bonds: 69.95%[34] - Commodities (excluding gold): 12.84%[34] - Gold: 8.9%[34] Next Month's Allocation Suggestion (June 2025) - **Model Allocation**: - Large Cap Stocks: 1.91%[35] - Small Cap Stocks: 0.96%[35] - Bonds: 88.54%[35] - Commodities (excluding gold): 2.79%[35] - Gold: 5.81%[35] - **Risk Allocation**: - Large Cap Stocks: 0.06[35] - Small Cap Stocks: 0.06[35] - Bonds: 4[35] - Commodities (excluding gold): 0.125[35] - Gold: 1[35]

Group 1 - The sudden drop in the Beijing Stock Exchange (BSE) stocks, with the BSE 50 index falling nearly 6%, is attributed to overcrowding in micro-cap stocks and significant overbought signals in the market [1][2] - The trading volume of the CSI 2000 has exceeded the high point from March, indicating a potential correction in the market [2] - The liquidity situation shows a contraction in net inflows for margin trading and a significant outflow from equity ETFs, leading to a shrinking demand side [2] Group 2 - The U.S. fiscal deficit is projected at 6.1% for this year, corresponding to approximately $2 trillion, with warnings from Treasury Secretary Yellen about the depletion of government cash and unconventional measures by August [2][3] - The global financial system is facing increased uncertainty due to the U.S. fiscal crisis and rising U.S. debt rates, prompting a shift towards "safe assets" [3] - The Japanese long-term interest rates have risen due to comments from Prime Minister Kishida, which may impact global market risk appetite, although the Bank of Japan may intervene if rates continue to rise significantly [3] Group 3 - A-shares have maintained a volatile trend since the second week of May, with a significant number of stocks declining, yet the Shanghai Composite Index remains strong [4] - Long-term expectations for quality assets in A-shares remain positive, with Morgan Stanley's target for the MSCI China Index set at 80 in a basic scenario and 89 in an optimistic scenario [4] - The Chinese economy's recovery is expected to rely on consumption and investment, with policy support needed to boost consumer spending [5]