金工策略周报-20260301

1. Report Industry Investment Rating There is no information provided regarding the report's industry investment rating. 2. Core Viewpoints of the Report - In the Treasury bond futures market, last week, all Treasury bond futures contracts ended higher, but the 30 - year, 10 - year, 5 - year, and 2 - year main contracts ended lower by 1.05%, 0.19%, 0.16%, and 0.06% respectively. The basis of each variety was differentiated. The market risk preference weakened, activating the hedging attribute of Treasury bond futures. The downward trend of Treasury bond futures is not likely to reverse when the long - term bull market logic of the stock market remains unchanged and the coupon income of Treasury bonds is not attractive. Only when the expected return of equity or risk assets declines marginally, the short - term hedging trading attribute of the bond market becomes more obvious. The medium - to - long - term trend strategy is still a relatively good choice for CTA strategies recently [5]. - After the Spring Festival, the domestic commodity market showed mixed trends. Popular varieties continued the high - volatility feature before the festival. Lithium carbonate, silver, and tin were the most significantly rising varieties last week, with an increase of over 10%, while polysilicon and caustic soda fell by over 5%. Overall, there were more rising varieties than falling ones. Among commodity factors, except for the basis and basis momentum factors which had negative returns, other types of commodity factors achieved positive returns in the cross - section. The volume - price trend factors had an average increase of over 1.5%, being the most outstanding factors last week [13][16]. 3. Summary by Relevant Catalogs 3.1 Treasury Bond Futures Quantitative Strategy 3.1.1 Treasury Bond Futures Market Review - Last week, all Treasury bond futures contracts ended higher, but the 30 - year main contract fell by 1.05%, the 10 - year main contract fell by 0.19%, the 5 - year main contract fell by 0.16%, and the 2 - year main contract fell by 0.06%. The basis of each variety was differentiated. The 10 - year CTD bond was 250025, and the basis on the 27th was about 0.03 yuan, slightly lower than the historical average. The 30 - year CTD bond was 210014, and the basis on the 27th was 0.43 yuan, slightly higher than the historical average [5]. - The market risk preference gradually weakened, activating the hedging attribute of Treasury bond futures. When the long - term bull market logic of the stock market remains unchanged and the coupon income of Treasury bonds is not attractive, the downward trend of Treasury bond futures is not likely to reverse. Only when the expected return of equity or risk assets declines marginally, the short - term hedging trading attribute of the bond market becomes more obvious [5]. 3.1.2 Treasury Bond Futures Daily - Frequency Timing Strategy - For the 10 - year Treasury bond, from 2021/01/01 to the present (out - of - sample), with a single - leverage portfolio, the annualized return, Sharpe ratio, and maximum drawdown were 2.76%, 1.31, and 2.13% respectively. Since the report was released (from 2025/11/01 to the present), with a single - leverage portfolio, the annualized return, Sharpe ratio, and maximum drawdown were 2.76%, 1.69, and 0.67% respectively [5]. 3.1.3 Unilateral Strategy Performance - The full - sample and post - report - release annualized returns were both 2.76%. The annualized volatility was 2.13% for the full sample and 1.63% after the report was released. The annualized Sharpe ratio was 1.30 for the full sample and 1.69 after the report was released. The maximum drawdown was 2.13% for the full sample and 0.67% after the report was released. The Calmar ratio was 1.30 for the full sample and 4.09 after the report was released [8]. 3.1.4 Strategy Details - The categories of large - scale factors include basis, intraday technical, intraday volume - price, high - frequency capital flow, member positions, and risk assets. The signals are generated by equal - weighting and averaging within large - scale factors, and the positive or negative sign of the average is used as the long - short signal. The strategy uses the VWAP of the first ten minutes of the next - day's opening as the trading price and buys with single leverage [11]. 3.2 Commodity CTA Factor and Strategy Performance 3.2.1 Commodity Factor Performance - After the Spring Festival, the domestic commodity market showed mixed trends. Popular varieties continued the high - volatility feature before the festival. Lithium carbonate, silver, and tin were the most significantly rising varieties last week, with an increase of over 10%, while polysilicon and caustic soda fell by over 5%. Overall, there were more rising varieties than falling ones. Among commodity factors, except for the basis and basis momentum factors which had negative returns, other types of commodity factors achieved positive returns in the cross - section. The volume - price trend factors had an average increase of over 1.5%, being the most outstanding factors last week, indicating that the post - festival market continued the previous price trend of bulk commodities, and the medium - to - long - term trend strategy is still a relatively good choice for CTA strategies recently [13][16]. 3.2.2 Tracking Strategy Performance | Strategy Name | Annualized Return | Sharpe Ratio | Calmar Ratio | Maximum Drawdown | Recent One - Week Return | Year - to - Date Return | | --- | --- | --- | --- | --- | --- | --- | | CWFT | 9.3% | 1.60 | 1.06 | - 8.81% | 0.96% | 1.59% | | C_frontnext & Short Trend | 11.2% | 1.72 | 1.67 | - 6.72% | 0.41% | 0.92% | | Long CWFT & Short CWFT | 12.2% | 1.38 | 0.94 | - 13.07% | 1.26% | 2.44% | | CS XGBoost | 5.0% | 0.82 | 0.24 | - 21.19% | - 1.15% | - 4.82% | | RuleBased TS Sharp - combine | 11.3% | 1.48 | 1.37 | - 8.26% | 0.42% | - 1.52% | | RuleBased TS XGB - combine | 11.3% | 1.98 | 2.53 | - 4.49% | 0.18% | - 1.34% | | CS strategies, EW combine | 12.6% | 1.81 | 1.71 | - 7.38% | 1.15% | 2.14% | [14] 3.2.3 Strategy Composition - CWFT portfolio: A simple composite portfolio of Carry, Warrant, Futurespot, and Trend factors. Factors within the same large category are equally weighted, and different large categories are compounded with weights of 5:2:2:1. - C_frontnext & Short Trend portfolio: While not changing the position direction of long - term spread factors, it tries to hedge the negative returns of short - term price fluctuations. C_frontnext is the main factor, and Short Trend (16 trend factors with a window parameter of 20 days) is the secondary factor. - Long CWFT & Short CWFT portfolio: The four types of factors C, W, Fs, and T are equally weighted and compounded within each category and then compounded with weights of 5:2:2:1. The long - cycle CWFT factor is the main factor, and the short - cycle CWFT factor is the secondary factor. - Cross - section CS XGB portfolio: A time - series factor XGB portfolio based on all market varieties, trained with samples from 20091231 - 20191231. - RuleBased TS Sharp - weighted portfolio: A time - series factor strategy based on rule - based long - short signals, obtained by stacking factor libraries with rules and coarsening window parameters, and then obtaining a Sharpe - weighted portfolio after in - sample screening. - RuleBased TS XGB portfolio: A time - series factor strategy based on rule - based long - short signals, obtained by stacking factor libraries with rules and coarsening window parameters, and then obtaining an XGBoost portfolio after in - sample screening [20]. 3.2.4 Strategy Position and Performance Details - CWFT strategy: Last week, it held 26 varieties, with a net position of 18.2%, a total position return of 1.0%, and a win - rate of 61.5%. This week, it holds 26 varieties, with a net position of 18.4%, 0 varieties need to be rolled over, and the total turnover capital ratio is 23.0% [22]. - C_frontnext & Short Trend strategy: Last week, it held 26 varieties, with a net position of 33.6%, a total position return of 0.4%, and a win - rate of 46.2%. This week, it holds 26 varieties, with a net position of 37.6%, 1 variety needs to be rolled over, and the total turnover capital ratio is 63.8% [24]. - Long CWFT & Short CWFT strategy: Last week, it held 26 varieties, with a net position of 58.2%, a total position return of 1.3%, and a win - rate of 69.2%. This week, it holds 26 varieties, with a net position of 56.3%, 0 varieties need to be rolled over, and the total turnover capital ratio is 25.0% [26]. - CS XGBoost strategy: Last week, it held 24 varieties, with a net position of 0.0%, a total position return of - 1.1%, and a win - rate of 50.0%. This week, it holds 24 varieties, with a net position of 0.0%, 1 variety needs to be rolled over, and the total turnover capital ratio is 60.9% [28]. - RuleBased TS Sharp - combine strategy: Last week, it held 44 varieties, with a net position of - 36.5%, a total position return of 0.8%, and a win - rate of 68.2%. This week, it holds 45 varieties, with a net position of - 26.6%, 2 varieties need to be rolled over, and the total turnover capital ratio is 39.7% [30]. - RuleBased TS XGB - combine strategy: Last week, it held 44 varieties, with a net position of - 49.9%, a total position return of 0.5%, and a win - rate of 61.4%. This week, it holds 45 varieties, with a net position of - 32.9%, 2 varieties need to be rolled over, and the total turnover capital ratio is 55.0% [32]. 3.2.5 Strategy Performance Comparison - Among the six strategies, Long CWFT & Short CWFT performed the best last week with a return of 1.26%, and also performed the best year - to - date with a return of 2.44%. - The equal - weighted composite strategy of the above cross - section strategies (equal - weighted weekly returns) has an annualized return of 12.6%, a Sharpe ratio of 1.81, a Calmar ratio of 1.71, a maximum drawdown of - 7.38%, a recent one - week return of 1.15%, and a year - to - date return of 2.14% [36].