Wing Cheung

I am an investment banking quant turned financial mathematician with 15 years of experience in the financial industry, specialising in quantitative finance research, executive education, and consulting, as well as financial technology (FinTech) innovation and entrepreneurship. I have authored more than 30 publications in quantitative finance and developed Unified Behaviour-Consistent Portfolio Theory, which has influenced hundreds of professional fund managers worldwide and led to multiple FinTech innovations, earning top FinTech awards in the UK, Hong Kong, Singapore, and China, with recognition from organisations such as Accenture, UBS, and Alibaba.

Previously, I served as Head of Portfolio Analytics at Lehman Brothers and Nomura International plc (London), where my team was ranked No. 1 in the 2010 Institutional Investor (II) European “Quantitative Research” Survey. During my time as a practitioner, I spoke at both professional and academic research conferences, taught at financial institutions and graduate schools, supervised PhD-level analysts, and served as a peer reviewer for academic journals and funding bodies.

I hold a PhD from the University of Cambridge, where I specialised in mathematical financial modelling.

Rigorous theoretical foundations and robust solutions to real-world practical challenges; investment decision-making rationale, philosophy, behaviour and ethics; high-dimention decision theory; philosophy of science; financial information processing; market modelling; holistic factor and risk management; portfolio theorisation and engineering; unified and cognitive investment process; cost-aware portfolio construction and trading; digital asset management (robo-advisor); asset management and WealthTech innovations; investor skill versus luck; transparency and regulation; regulation and RegTech innovations; artificial intelligence (AI) versus science; harmonising financial models and theories with human cognition; risk-neutral pricing etc.

Intellectual Contributions

Finance is widely recognised as both an art and a science. If this is true, then the scientific aspect of finance should be governed by universal, elegant scientific principles. Accordingly, a central focus of my research is the theoretical development and empirical validation of a Unified Behaviour-Consistent Portfolio Theory — a parsimonious set of principles that explain, unify, and guide portfolio selection.

Since the publication of Modern Portfolio Theory (MPT) (Markowitz, 1952), practitioners and academics alike (e.g., Treynor and Black, 1972) have acknowledged its fundamentally passive nature and its disconnect from real-world investor behaviours. They have long called for an active portfolio theory that aims to outperform the market while more accurately reflecting real-world practices. My ABL model (Cheung, 2013) responds to this call by incorporating comprehensive, natural investor opinions, optimally translating investor forecasts into portfolio outperformance, and unifying diverse investor behaviours into a single framework. ABL has been recognised in both industry and academia for its ability to generate purer, more robust, and more diversified portfolios that remain efficiently aligned with investors’ intentions.

Using a novel Mechanism Dissection methodology, Cheung (2024a) derives a theoretical framework for active portfolio selection built on only two orthogonal principles — the Subjective Allocation Rule (SAR) and Minimum Tracking Error (MTE) — and demonstrates that this parsimonious set of principles jointly unifies theoretical modelling with real-world portfolio practice.

Cheung (2024b) rigorously validates the framework’s internal efficiency. Through a new Skill-based View Simulation (SVS) methodology, it is shown that SAR generates outperformance proportional to an investor’s forecasting skill. By designing a Comparative Subjectivity Testing (CST) framework, Cheung (2024b) further demonstrates that SAR can leverage investor cognition and subjectivity to enhance portfolio performance.

Cheung (2024c) empirically identifies the conditions under which investor subjectivity and alternative behaviours outperform MPT (i.e., mean–variance optimisation), drawing on the Matrix Perturbation Technique (PTT), which — together with SVS — overcomes the common joint-test problem in the portfolio theory validation literature.

Cheung (2025) (a working paper) advances the argument from parsimony to completeness, establishing an axiomatic foundation proving how SAR and MTE jointly deliver the explanatory, unificatory, and guiding powers expected of a coherent scientific framework. The outcome is a Unified Portfolio Theory.

As a unified theory, it underpins numerous practical applications, leading to the award-winning FinTech invention of Universal Portfolio Optimiser (UPO), Real-Time Personalisable Robo-Advisor (RA2.0), Custom Factor Analytics (CFA), Fund-of-Funds (FoF) Optimisation, Holistic Factor Management (HFM), and Decision Performance Attribution (DPA).

Looking ahead, there is considerable scope for collaborative exploration — including evaluation across multiple asset classes, varying market regimes, and alternative investor heuristics — to further test the robustness and generalisability of SAR and MTE. Extensions incorporating non-normality, behavioural biases, and machine-learning-generated signals also present rich opportunities for joint research. From an applied perspective, user-centric decision support tools — such as interactive dashboards and AI-assisted portfolio massage — could be co-developed to operationalise these principles in real-time portfolio management and advisory, bridging theoretical advances with practice. I warmly welcome collaboration from fellow academics and PhD researchers on any of these forthcoming directions.

SSRN Author page:

https://papers.ssrn.com/sol3/cf_dev/AbsByAuth.cfm?per_id=1171776