Research: Evaluating Portfolio Optimisation in Robo-Advisory Platforms with Heterogeneo...

Research question: How do robo-advisory portfolio optimisation algorithms perform in tailoring asset allocation to heterogeneous client risk preferences compared to traditional advisory models?

Background and Motivation: The rapid adoption of robo-advisors in wealth management has transformed portfolio construction, offering automated, algorithm-driven investment solutions accessible to a broad client base. These platforms promise the ability to tailor portfolios to individual risk preferences efficiently and at scale, leveraging advanced optimisation techniques.

Research Gap: While prior studies have assessed the general performance of robo-advisors, there is limited empirical research on how well their optimisation algorithms accommodate the diversity of client risk profiles compared to traditional, human-advised portfolios. The effectiveness of these algorithms in reflecting nuanced risk preferences remains underexplored.

Approach and Expected Contribution: This project will systematically review the algorithms used in leading robo-advisory platforms, develop an experimental framework using simulated and real client profiles, and quantitatively compare the resulting portfolios against those constructed via traditional advisory benchmarks. Statistical and econometric methods will be applied to assess alignment with stated risk preferences, risk-adjusted returns, and downside protection.

Significance: Understanding the efficacy and limitations of robo-advisory optimisation in capturing client heterogeneity is crucial for investors, regulators, and platform designers. The findings will inform best practices and could influence the development of next-generation digital advisory models.