Home

COURSE AIMS & OBJECTIVES, KEY SKILLS AND LEARNING OUTCOMES

Course Aims & Objectives

Artificial Intelligence (AI) and Machine Learning (ML) increasingly encroach every corner of our lives. The course will help participants understand how AI differs from ML, in particular, how ML is disrupting the landscape of business and finance. Course participants are expected to develop the confidence and the skills needed to define problems in business and finance, and to tackle them using Python ML toolkits.

Desired Outcomes: on successful completion, participants should be able to:

  1. Understand the difference between AI and ML.
  2. Familiarize themselves with ML Models.
  3. Understand and appreciate the impact of ML, particularly in financial markets.
  4. Understand the basics of preparing and working with data for ML applications.
  5. Discover how the Python toolkit can be used in ML.
  6. Build an ML model using Python.

Learning Method & Delivery Format

This course includes lectures, readings and coding. All material will be available in the course Dropbox file.

All participants are required to participate actively in class discussions. They must be prepared to address questions and issues arising from their work and to discuss related issues, other points of view and opposing ideas.

In daily 6-hour sessions over 3 days, the course will be delivered online via either Zoom or Teams. Each interactive session will provide a combination of discussion and hands-on applications.

Handouts include Course notes, PDF, pptx and Python code.

Course Prerequisites

Although some preliminary Basic Python and financial knowledge is preferable, it is not a must. All participants are required to have Python already installed on their machines and Jupyter Notebook as a working environment; a link will be provided prior to the course.

Who Should Attend?

  • Postgraduate and Doctoral Students.
  • Early Career Researchers and Practitioners.
  • Traders.
  • Strategists.
  • Risk Managers.
  • Consultants.

Course Content

DAY 1: Machine Learning Fundamentals and Applications

  • An overview of ML. Artificial Intelligence (AI) vs Machine Learning (ML).
  • Most common types of ML approaches (supervised learning, unsupervised learning, and reinforcement learning). Different approaches to different types of problems.
  • The Artificial Neural Network (ANN).
  • Exploration of Linear Regression, Logistic Regression, Decision Tree, Random Forest, Support Vector Machines, Naïve Bayes, KNN, Deep Learning, Natural Language Processing and Reinforcement Learning at a granular level.
  • Synergies between ML and finance with emphasis on financial markets.
  • Applications of ML in financial markets.
  • Case Study: Robo-advisors.

DAY 2: Python ML toolkits and Data Management

  • Existing ML tools (platforms; open-source programming languages: Python, R).
  • Exploration of Python development environments.
  • Popular Python ML toolkits (Scikit-learn, Keras, Tensorflow, PyTorch).
  • Understanding the cruciality of data to ML.
  • Types of data (structured vs unstructured).
  • Learning how to prepare data according to the problem at hand. Feature Scaling.
  • Data management in Python I: pre-processing data, handling missing data and working with categorical data.
  • Data management in Python II: dividing data between a training set and a test set.

DAY 3: Practical Sessions – LSTM in the Financial Markets

  • The Long Short-Term Memory (LSTM) Model.
  • Strengths and Drawbacks of the LSTM Model.
  • The LightGBM Model.
  • The Word2Vec Model.
  • Set-up the Python development environment.
  • Two different applications with different datasets and different methodologies:
  • Practical Session I: Combining LSTM with LightGBM for Stock Price Prediction.
     Describing the task and outlining the methodology.
     Preparing and pre-processing the dataset.
     Building an LSTM + LightGBM model for stock price prediction using Python.
     Training/testing the ML algorithm and making predictions.
     Analysing and discussing the results.
  • Practical Session II: Combining LSTM and Word2Vec for Sentiment Analysis
     Describing the task and outlining the methodology.
     Visualising the data.
     Preparing and pre-processing the data.
     Build an LSTM + Word2Vec model for sentiment analysis using Python.
     Training/testing the ML algorithm using training/testing data sets and user-generated data.
     Analysing and discussing the results.

Course Instructor

As a quantitative financial researcher, Dr Jolnar Assi has held many roles in the City of London and internationally. Her rich experience in banking builds upon roles in London with JP Morgan and TD Securities, engaging with equity derivatives, risk measurement, fixed income derivatives and credit hybrids. She has briefly served as a risk manager at Emirates NBD in Dubai. With her work towards an MBA and a PhD, she has acquired experience of Artificial Intelligence (AI) and Machine Learning (ML). Most recently, Jolnar has focused on AIML applications and building her own e-learning platform: Traders Island Ltd.

Jolnar Assi – LinkedIn

Prof Marwan Izzeldin
Director: GOLCER
November 2022