Robust Risk Parity and Covered Call Optimization

稳健的风险平价和备兑看涨期权优化

• 批准号: RGPIN-2019-05733
• 负责人: Kwon, Roy
• 金额: $ 2.26万
• 依托单位: University of Toronto
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2021
• 资助国家: 加拿大
• 起止时间: 2021-01-01 至 2022-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=738481
• 关键词: Robust; Risk; Parity; Covered; Call

The proposed research will focus on the development of mathematical models and algorithms for financial investment design. We consider investment (asset allocation) frameworks based on risk parity optimization, index tracking, and covered call overlays. Risk parity is a relatively new approach that uses mathematical optimization to create investment decisions where risk contributions from individual assets are equal which readily results in diversified portfolios unlike the use of mean-variance optimization. However, little attention has been given to determine how robust risk parity models are. We have found that naïve approaches for risk parity optimization can exhibit considerable sensitivity to parameter estimations. The development of a robust formulation of risk parity optimization is a major goal of our research. We will incorporate expected returns and short selling to form general risk-return risk parity-based models. The benefit is that an investor will have the ability to better trade-off return and degree of risk parity in constructing investment strategies. Portfolio indexing is a passive investing strategy that aims to replicate the risk and return profile of a broad market index such as the S&P 500. Indices like the S&P 500 have performed well and studies have shown that most active managers that use market timing or stock picking do not outperform the indices like the S&P 500. However, market indices like the S&P 500 have some serious limitations due to their market capitalization weightings that have motivated alternative approaches. We will design portfolios to be reasonably close to a market index but limits the risk concentration of the assets by using risk parity. Expected return estimations are then required, which we will mitigate estimation error by considering robust optimization. Option overlays such as covered call writing has emerged ras an effective method of enhancing returns of a portfolio. This is where an investor will sell call options on the assets of her portfolio. One limitation of this strategy is that call options are sold in correspondence to the entire position of an asset. Also, covered call overlays also assume that the investment portfolio is known prior to selling the call options. We seek to develop optimization models to capture a generalized cover call problem where both investment allocation decisions and covered call selling are done simultaneously under uncertainty in asset prices. As a special case, simultaneous risk parity and covered call optimization will be developed. The results from the proposed research will extend the field of financial optimization and have significant practical benefits in the financial investment industry which is an important sector of the Canadian economy. The training of HQP in the research program will prepare them to engage in industry as quantitative financial professionals adding strength to Canada's vibrant financial industry.

拟议的研究将侧重于金融投资设计的数学模型和算法的开发，我们认为基于风险平价优化、指数跟踪和备兑看涨期权叠加的投资（资产配置）框架是一种相对较新的使用数学的方法。与使用均值方差优化不同，我们发现单个资产的风险贡献相等，因此可以通过优化来创建投资决策，这很容易导致投资组合多样化。但是，我们发现这种方法的稳健性却很少。风险平价优化可以对参数估计表现出相当大的敏感性。我们研究的一个主要目标是开发稳健的风险平价优化公式，以形成基于风险平价的一般风险回报模型。投资组合指数是一种被动投资策略，旨在复制标准普尔 500 指数等广泛市场指数的风险和回报状况。指数如标准普尔 500 指数表现良好，研究表明，大多数利用市场时机或选股的主动型基金经理的表现并不优于标准普尔 500 等指数。然而，标准普尔 500 等市场指数由于其市值我们将设计合理地接近市场指数的投资组合，但需要使用风险平价来限制资产的风险集中度，我们将通过这种方式减少估计误差。考虑到提高投资组合回报的有效方法，诸如备兑看涨期权之类的期权覆盖已经出现，这是投资者出售其投资组合资产的看涨期权的一个限制。此外，备兑看涨期权覆盖还假设投资组合在卖出看涨期权之前是已知的，我们寻求开发优化模型来捕获广义的备兑看涨期权问题，其中投资分配决策和投资组合都是已知的。在资产不确定的情况下同时进行备兑看涨期权卖出作为一个特例，将开发同步风险平价和备兑看涨期权优化，所提出的研究结果将扩展金融优化的领域，并对作为加拿大经济重要部门的金融投资行业产生重大的实际效益。 HQP 在研究项目中的培训将使他们做好准备，作为金融专业人士参与行业，为加拿大数量上充满活力的金融业增添力量。