在全球经济深度融合与市场波动频繁的背景下,行业资产配置成为投资者、金融机构及学术界关注的焦点。本文聚焦于基于改进的Black-Litterman模型的行业资产配置实证研究,通过引入ARMA模型对Black-Litterman模型的观点收益率进行预测优化,构建了一个更为精准和实用的行业资产配置模型。研究选取了2023年1月3日至2024年12月31日的日度数据,涵盖金融、消费、信息技术、工业及医药五大行业的代表性股票。实证结果表明,改进的Black-Litterman模型在风险控制和收益提升方面表现出色,能够有效提高投资组合的累计收益率,增强投资策略的稳健性和适应性。具体来看,模型优化后的资产配置方案使投资组合在市场波动时能够保持相对稳定的收益,同时在长期投资中有效平衡风险与收益。本研究为投资者提供了科学合理的资产配置方案,有助于在复杂多变的市场环境中优化行业资产配置,实现资产的稳健增值。同时,本研究也为金融机构提供了更为精准和个性化的资产配置服务,具有重要的实践意义和理论价值。In the context of today’s deeply integrated global economy and frequent market fluctuations, industry asset allocation has become a focal point for investors, financial institutions, and the academic community. This paper focuses on an empirical study of industry asset allocation based on an improved Black-Litterman model, introducing an ARMA model to predict and optimize the view returns of the Black-Litterman model, thus constructing a more precise and practical industry asset allocation model. The study selected daily data from January 3, 2023, to December 31, 2024, covering representative stocks in five major industries: finance, consumer goods, information technology, industry, and pharmaceuticals. The empirical results show that the improved Black-Litterman model excels in risk control and return enhancement, effectively increasing the cumulativ