DATA-DRIVEN MODELS FOR INVESTMENT

DATA-DRIVEN MODELS FOR INVESTMENT

Antonio Simeone, Villy Edoardo de Luca

Instructional goals

•Learn how to use alternative data like satellite and geospatial data, web data and news/media with LLM and Generative AI to improve investment decisions •Explore Machine Learning and Decision-Making methods with a focus on investment applications. •Implement a live trading strategy following the most valuable Hedge Fund techniques •Build a systematic trading strategy with both a qualitative and a quantitative AI approach •Agentic AI and Transformer-based architectures for forecasting financial time series.

Intended learning outcomes

Qualitative AI: Understand the principles of Generative AI and Large Language Models, with a focus on financial applications. Analyze the evolution of NLP, from rule-based systems to Transformers, LLMs and AI Agents. Apply Transformer-based architectures and multi-agent systems to financial use cases. Quantitative AI: Integrate Fuzzy Logic and Neural Network to predict financial time series Apply Transformer foundation models to financial time series forecasting. Develop a systematic trading strategy integrating qualitative and quantitative AI.

Course Contents

Evolution of Artificial Intelligence in Finance: from rule-based NLP to LLMs and AI Agents. Financial AI use cases: FinBERT, AlphaAgents, TradingAgents. Generative Models for Financial Time Series. Foundation Models for Time Series: TimeGPT, Chronos, TabPFN. Real & Complex Fuzzy Logic Adaptive Fuzzy Neural Networks Multi-Agent Systems for financial decision-making. Systematic Trading.

Reference Books

The course materials will be based on collection of top scientific articles and book reports. Articles and other materials (cases) will be available through the myluiss platform.

Teaching Methods

Interactive lectures Real time coding sessions Teamwork Project work

Assessment Method

Attending compliant students (presence >= 70%): Mid-term project, regarding the creation of a systematic trading strategy integrating qualitative and quantitative AI approaches, evaluated up to 11 points. Final oral exam evaluated up to 20 points. Each vote greater than 30 will be 30L. Non attending students or retake session: Oral exam evaluated up to 30L

Thesis assignment criteria

The thesis will be assigned (upon specific request to the instructor) to students who demonstrate a serious and motivated interest to the course subjects.

Week 1

Introduction to the course. Systematic trading. Practical Lab Session with Python.

Week 2

From Natural Language Processing to Large Language Models, financial use cases. Practical Lab Session with Python.

Week 3

Collection processing and Analysis of Alternative Data for financial usage. Practical Lab Session with Python

Week 4

Bloomberg terminal overview. Practical Lab Session with the terminal

Week 5

Foundation Models for Financial Time Series: TimeGPT (Nixtla) and Forecasting as a Service, Chronos (Amazon Science) with T5-based tokenization and quantization, TabPFN for structured data reasoning. Zero-shot inference on unseen data. Practical Lab Session with Python.

Week 6

AI Agents in Finance. From static LLMs to autonomous reasoning entities integrating tools, memory and interaction loops. The agent loop: Perceive, Reason, Act, Observe. Multi-agent systems for equity portfolio construction and trading (AlphaAgents, TradingAgents). Practical Lab Session with Python.

Week 7

Learning and application real techniques deployed in one of the most important UK based Hedge Fund. Overview of the Hedge Fund industry. Insights from the Head of Capital Markets at this Hedge Fund

Week 8

Quantum Annealing Introduction to spin networks. Classical/quantum Ising models.

Week 9

Applications.Mapping and embedding of an optimization problem onto a spin network. Practical example and application on financial services.

Week 10

Integration of Qualitative and Quantitative AI, ensemble models. Practical Lab Session with Python

Week 11

Genetic Algorithms. Practical Lab Session

Week 12

Recap and Q&A Practical Lab Session with Python