<link>https://don-huang.com/</link></image><item><title>Introduction of AI Quant Investment</h1> <article> <h1>Introduction of AI Quant Investment</h1> <p>Sun, 19 Jul 2026 00:00:00 +0000</p> <h2 id="课程简介">课程简介</h2> <p>本次线上直播课程面向个人投资者，以"别和机器拼手速"为主题，系统讲解如何借鉴量化投资思维、识别行为误区、善用 AI 工具来提升投资决策质量。课程分为三大模块：第一模块介绍量化投资的基本框架与六类经典投资因子（价值、质量、动量、低波动、规模、短期反转），结合 A 股真实案例讲解因子的适用场景、失效风险与回测陷阱；第二模块通过六个互动测验，剖析前景理论、处置效应、锚定效应、过度自信、羊群效应等行为偏误如何影响交易决策，并介绍 Pre-mortem 等纪律工具；第三模块现场演示 AI Agent 在投资研究中的应用，包括数据质量审查、因子解释与冲突分析、回测审计、多角色投资委员会模拟，以及 Human-in-the-loop 决策流程。</p> <h2 id="course-overview">Course Overview</h2> <p>This two-hour online lecture, themed “Don’t Race Against Machines,” equips individual investors with a scientific, reproducible investment decision framework. Module 1 introduces quantitative investing fundamentals and six classic factor families (value, quality, momentum, low-volatility, size, and short-term reversal), illustrated with real A-share market cases covering factor applicability, failure risks, and backtest pitfalls. Module 2 uses six interactive quizzes to dissect behavioral biases—prospect theory, disposition effect, anchoring, overconfidence, and herding—and presents disciplinary tools such as Pre-mortem analysis. Module 3 demonstrates AI Agent applications in investment research, including data quality auditing, factor interpretation and conflict analysis, backtest review, multi-role investment committee simulation, and Human-in-the-loop decision workflows.</p> <h2 id="资料下载">资料下载</h2> <ul> <li> </li> <li> </li> </ul> </article> <article> <h1>Live Lecture: Introduction of AI Quant Investment</h1> <p>Sun, 19 Jul 2026 00:00:00 +0000</p> <p>Delivered a two-hour online live lecture themed “Don’t Race Against Machines,” covering quantitative investment factors, behavioral biases in trading, and practical AI Agent demonstrations for investment research and decision-making.</p> <p> </p> </article> <article> <h1>AI Application in FinTech</h1> <p>Fri, 17 Jul 2026 00:00:00 +0000</p> <h2 id="课程简介">课程简介</h2> <p>本课程面向具有基础数理背景的学生，系统介绍人工智能在金融科技领域的技术原理与实际应用。课程分为两大部分：第一部分从图灵测试出发，梳理 AI 技术的演进脉络，涵盖监督学习、无监督学习、强化学习三大范式，深入讲解神经网络、注意力机制、Transformer 架构、大语言模型（LLM）以及检索增强生成（RAG）和 AI Agent 等前沿概念。第二部分聚焦金融科技应用场景，包括智能投顾（Robo-Advisory）、组合优化与现代投资组合理论、直接指数化（Direct Indexing）、税收亏损收割（Tax-Loss Harvesting），以及如何将多角色 AI Agent 系统应用于投资研究与决策流程。课程通过真实案例和实操演示，帮助学生理解如何将机器学习与大模型能力转化为可部署的金融产品与服务。</p> <p> <figure > <div class="flex justify-center "> <div class="w-full" ></div> </div></figure> </p> <h2 id="资料下载">资料下载</h2> <ul> <li> </li> <li> </li> </ul> </article> <article> <h1>Guest Lecture: AI Application in FinTech at HKUST</h1> <p>Fri, 17 Jul 2026 00:00:00 +0000</p> <p>Delivered a guest lecture at the Hong Kong University of Science and Technology as part of the HKUST Lecture Series. The talk covered AI technology foundations—from machine learning paradigms to large language models and AI agents—and their applications in FinTech, including robo-advisory, direct indexing, and agentic investment workflows.</p> <p> </p> </article> <article> <h1>A Two-Layer Attribution Framework for Portfolios Built from Mixed-Asset Products</h1> <p>Thu, 01 Jan 2026 00:00:00 +0000</p> </article> <article> <h1>Drift Minimization under Lot-Size Constraints: A Practical Integer-Share Rebalancing Problem in Robo-Advisory</h1> <p>Wed, 01 Jan 2025 00:00:00 +0000</p> </article> <article> <h1>AI-Driven Automatic Wealth Management Solution</h1> <p>Sun, 31 Mar 2024 00:00:00 +0000</p> <p><strong>Period:</strong> April 2021 – March 2024<br> <strong>Funding:</strong> Innovation and Technology Fund (ITF), Hong Kong, Reference No. B/E003/20<br> <strong>Industry partner:</strong> Magnum Research Limited (AQUMON)</p> <h3 id="problem">Problem</h3> <p>End-to-end digital wealth advisory requires integrating client profiling, goal-based portfolio construction, ongoing rebalancing, and monitoring within a scalable, auditable, and compliance-aware system. Existing approaches either rely on static rule-based logic or lack the modularity required for institutional deployment across diverse client segments and regulatory jurisdictions.</p> <h3 id="method">Method</h3> <p>The project developed a full-stack AI advisory architecture covering: (i) client profiling and suitability assessment using ML classifiers; (ii) goal-based portfolio construction integrating intertemporal allocation theory with modern estimation techniques; (iii) a hybrid-cloud data foundation supporting global asset universes and real-time feeds; (iv) high-fidelity backtesting with controls against overfitting; and (v) integration with execution infrastructure. Research components addressed model selection, uncertainty quantification in portfolio optimisation, and the design of modular research-validation-deployment pipelines.</p> <h3 id="outcome">Outcome</h3> <p>The project delivered a production-grade automated wealth management architecture deployed across multiple institutional client implementations. It established a reusable technical foundation for AQUMON’s robo-advisory and AI advisory services.</p> </article> <article> <h1>AI-Driven MPF Retirement Advisory System</h1> <p>Mon, 01 Jan 2024 00:00:00 +0000</p> <h3 id="problem">Problem</h3> <p>Hong Kong’s Mandatory Provident Fund (MPF) system requires participants to actively select from a range of constituent funds within their employer-sponsored scheme, yet most participants lack the financial knowledge or time to make well-informed, dynamic fund-selection decisions. Retirement adequacy outcomes are systematically poor when fund allocation is not regularly reviewed.</p> <h3 id="method">Method</h3> <p>AQUMON developed an AI-driven MPF investment advisory system combining algorithmic goal-based personalisation, lifecycle and asset-liability management (ALM) informed optimisation, and explainable recommendation workflows. The system integrates personalised retirement-gap analytics, risk profiling, and constraint-aware fund selection within a compliance-ready architecture suitable for deployment in regulated insurance and pension distribution environments. Distributed computation enables real-time recommendation generation across large participant bases.</p> <h3 id="deployment">Deployment</h3> <p>The system has been deployed with multiple MPF trustees and insurance providers in Hong Kong, including AIA (integrated into the AIA+ mobile app) and BCT Group (MARIO, BCT’s market-first AI advisor). The platform is designed to support approximately seven million Hong Kong MPF participants, with cumulative assets under advice exceeding USD 10 billion.</p> <h3 id="research-connection">Research connection</h3> <p>This project draws on doctoral and post-doctoral research in stochastic control and optimal withdrawal policy design. The production deployment extends those theoretical results into a regulated, real-world context.</p> </article> <article> <h1>Defining and Measuring Portfolio Health: A Drift-Based Metric Relative to Model Portfolios</h1> <p>Mon, 01 Jan 2024 00:00:00 +0000</p> </article> <article> <h1>Experience</h1> <p>Tue, 24 Oct 2023 00:00:00 +0000</p> </article> <article> <h1>ESG Quantitative Scoring System and ML-Driven Investment Research</h1> <p>Sun, 01 Jan 2023 00:00:00 +0000</p> <h3 id="problem">Problem</h3> <p>Existing ESG ratings from major providers suffer from low cross-provider agreement, limited transparency, and methodological subjectivity. A more systematic, data-driven approach is needed that integrates structured financial data with unstructured news and disclosure content, updates at high frequency, and supports both index construction and active strategy development.</p> <h3 id="method">Method</h3> <p>AQUMON developed a four-layer evaluation framework — Pillar, Category, Topic, and Metric — covering environmental, social, and governance dimensions for a universe of listed companies. The system integrates structured financial and regulatory data with unstructured news, corporate disclosures, and sentiment signals using multimodal ML pipelines. Dynamic weight updating and cross-industry comparability were explicit design requirements. The scoring methodology was designed to satisfy IFRS S1 and IFRS S2 disclosure frameworks and supports both institutional investment mandates and corporate sustainability reporting.</p> <h3 id="outcome">Outcome</h3> <p>The system provides daily ESG score updates covering 3,000+ listed companies. Backtested long-only strategies combining ESG scores with fundamental signals demonstrated improved risk-adjusted performance relative to market benchmarks. The framework has been adopted by the HKSAR Government and multiple listed companies for sustainability reporting and investment mandate compliance.</p> </article> <article> <h1>Portfolio Management via Reinforcement Learning</h1> <p>Fri, 31 Dec 2021 00:00:00 +0000</p> <p><strong>Period:</strong> January 2019 – December 2021<br> <strong>Funding:</strong> Innovation and Technology Fund (ITF), Hong Kong, Reference No. UIM/365<br> <strong>Industry partner:</strong> Magnum Research Limited (AQUMON)</p> <h3 id="problem">Problem</h3> <p>Standard portfolio rebalancing rules are based on fixed-weight targets and threshold-trigger mechanisms that do not adapt to evolving market conditions or explicitly account for transaction costs. A principled sequential decision-making framework is needed to learn rebalancing policies that remain effective under diverse market regimes and realistic cost structures.</p> <h3 id="method">Method</h3> <p>The portfolio rebalancing problem is formulated as a Markov Decision Process (MDP) with state variables capturing portfolio composition, market features, and risk indicators; actions corresponding to rebalancing decisions; and a reward function that penalises both tracking error and transaction costs. We explored value-based and policy-gradient reinforcement learning algorithms, including recurrent and sequence-based architectures that condition on the trajectory of market states. Emphasis was placed on sample efficiency, robustness to distribution shift, and deployability in a production rebalancing pipeline.</p> <h3 id="outcome">Outcome</h3> <p>The project produced a research-to-production foundation for RL-based portfolio management, with documented methodology, experimental benchmarks, and a reusable framework for systematic strategy development and evaluation at AQUMON.</p> </article> <article> <h1>Optimal Initiation of Guaranteed Lifelong Withdrawal Benefit with Dynamic Withdrawals</h1> <p>Sun, 01 Jan 2017 00:00:00 +0000</p> </article> <article> <h1>Weak Convergence of Path-Dependent SDEs and Functionals in Pricing Basket CDS with Counterparty Risk and Contagion Risk</h1> <p>Sun, 01 Jan 2017 00:00:00 +0000</p> </article> <article> <h1>Regression-Based Monte Carlo Methods for Stochastic Control Models: Variable Annuities with Lifelong Guarantees</h1> <p>Fri, 01 Jan 2016 00:00:00 +0000</p> </article> <article> <h1>Regression-Based Monte Carlo Methods for Stochastic Control Models</h1> <p>Thu, 13 Nov 2014 00:00:00 +0000</p> </article> <article> <h1>Analysis of Optimal Dynamic Withdrawal Policies in Withdrawal Guarantee Products</h1> <p>Mon, 07 Jul 2014 00:00:00 +0000</p> </article> <article> <h1>Analysis of Optimal Dynamic Withdrawal Policies in Withdrawal Guarantee Products</h1> <p>Mon, 02 Jun 2014 00:00:00 +0000</p> </article> <article> <h1>Analysis of Optimal Dynamic Withdrawal Policies in Withdrawal Guarantee Products</h1> <p>Wed, 01 Jan 2014 00:00:00 +0000</p> </article> <article> <h1>Analysis of Optimal Dynamic Withdrawal Policies in Withdrawal Guarantee Products</h1> <p>Thu, 26 Sep 2013 00:00:00 +0000</p> </article> </main></body></html>