证券研究报告|金融工程动态报告 金融工程 2026年03月02日 长期继续看多黄金，短期关注原油和铜 ——另类投资策略周度跟踪 证券分析师： 王程畅 执业证书编号：S0210526020002 李杨 执业证书编号：S0210524100005 华福证券 摘要 2 华福证券 华福证券 ➢ 核心观点：A股情绪指数抬升，港股情绪指数下降，上证50、沪深300、中证500和中证1000的VIX下降，对A股的择时 仓位为看多，对港股的择时仓位为中性。当前机构关注基础化工和汽车行业，非银行金融行业的机构关注度从高位下降 。最近一周"煤炭"、"电力及公用事业"、"银行"、"非银行金融"和"传媒"行业的机构关注度在提升。当前石 油石化、有色金属、钢铁、基础化工和建材行业处于触发拥挤指标阈值的状态（流动性、成分股扩散、波动率）。 2026年03月看好传媒、电子、汽车和农林牧渔的相对收益。黄金和白银的VIX高位下降，铜和原油的VIX高位震荡。美 国实际利率下行、市场波动增加、地缘政治风险抬升、黄金需求增加，中长期依然看多黄金。 ➢ A股和港股投资者情绪跟踪：A股方面，"A股情绪指数_等权"抬升，上证50、沪深300、中证500和 ...

- Model Name: CSI 500 Enhanced Portfolio; Model Construction Idea: The model aims to outperform the CSI 500 index by selecting stocks with higher expected returns based on quantitative strategies[2][58] - Model Construction Process: The model uses a quantitative strategy to select stocks from the CSI 500 index. The portfolio's performance is evaluated based on its excess return over the CSI 500 index. The specific construction process involves selecting stocks with higher expected returns and adjusting the portfolio weights accordingly[58][61] - Model Evaluation: The model has shown a significant excess return over the CSI 500 index, indicating its effectiveness in enhancing returns[58][61] - Model Name: CSI 300 Enhanced Portfolio; Model Construction Idea: The model aims to outperform the CSI 300 index by selecting stocks with higher expected returns based on quantitative strategies[2][65] - Model Construction Process: The model uses a quantitative strategy to select stocks from the CSI 300 index. The portfolio's performance is evaluated based on its excess return over the CSI 300 index. The specific construction process involves selecting stocks with higher expected returns and adjusting the portfolio weights accordingly[65][66] - Model Evaluation: The model has shown a significant excess return over the CSI 300 index, indicating its effectiveness in enhancing returns[65][66] - Factor Name: Value Factor; Factor Construction Idea: The value factor aims to capture the excess returns of stocks that are undervalued relative to their fundamentals[2][70] - Factor Construction Process: The value factor is constructed by ranking stocks based on their valuation ratios, such as price-to-book (P/B) and price-to-earnings (P/E) ratios. Stocks with lower valuation ratios are considered undervalued and are given higher weights in the factor portfolio[70][76] - Factor Evaluation: The value factor has shown high excess returns, indicating its effectiveness in capturing the returns of undervalued stocks[70][76] - Factor Name: Residual Volatility Factor; Factor Construction Idea: The residual volatility factor aims to capture the excess returns of stocks with lower idiosyncratic risk[2][70] - Factor Construction Process: The residual volatility factor is constructed by ranking stocks based on their residual volatility, which is the volatility of the stock's returns unexplained by market movements. Stocks with lower residual volatility are given higher weights in the factor portfolio[70][76] - Factor Evaluation: The residual volatility factor has shown high excess returns, indicating its effectiveness in capturing the returns of low-risk stocks[70][76] - Factor Name: Non-linear Size Factor; Factor Construction Idea: The non-linear size factor aims to capture the excess returns of stocks with specific size characteristics that are not linearly related to market capitalization[2][70] - Factor Construction Process: The non-linear size factor is constructed by ranking stocks based on their non-linear size characteristics, which may include measures such as the square or cube of market capitalization. Stocks with specific size characteristics are given higher weights in the factor portfolio[70][76] - Factor Evaluation: The non-linear size factor has shown significant negative excess returns, indicating its ineffectiveness in capturing the returns of stocks with specific size characteristics[70][76] Model Backtest Results - CSI 500 Enhanced Portfolio, Excess Return: 46.94%, Maximum Drawdown: -4.99%[58][61] - CSI 300 Enhanced Portfolio, Excess Return: 31.61%, Maximum Drawdown: -5.86%[65][66] Factor Backtest Results - Value Factor, Excess Return: High[70][76] - Residual Volatility Factor, Excess Return: High[70][76] - Non-linear Size Factor, Excess Return: Significant Negative[70][76]