Programme

Programme

Programme

Day 1 - Risk

08:30

Registration 

09:00

Machine Learning Models

  • Supervised Vs Unsupervised models
    • Classification
    • Clustering
    • Regression
    • Dimensionality Reduction
  • Reinforcement Learning models
  • Deep Learning models
  • Selection of model
  • Explainability of model
  • Challenges, Ethics and Regulations of ML models

Harsh Prasad, Vice President, MORGAN STANLEY 

11:00

Morning coffee break

11:30

Understanding and developing an effective big data strategy

  • Extracting value from limited information
  • Three key components of big data analytics
  • Big data management: infrastructure and technology
  • Data availability, accessibility and integrity
  • Batch and stream processing
  • Advanced analytics: insights generation and algorithms development
  • Modern data analysis: structured/unstructured data modelling
  • Insights consumption: application and visualization

Johnson Poh, Head Data Science for Big Data Analytics Center of Excellence, DBS BANK

13:30

Lunch 

14:30

Risk Management via Alternate Data Sources

  • Identifying entities, relations and other metadata from unstructured data
  • Hands on python exercises for tagging unstructured data
  • Understanding graph data - rdf
    • Converting relational data to graph data
  • Tagging news items and getting real time sentiment
    • Use case demo and discussion
  • Impact of news on portfolio of securities
  • Supply chain risk management
    • Use case demo and discussion

15:30

Afternoon coffee break 

16:00

Machine learning in banking, risk management & modelling

  • ML applications in banking and risk management
  • Analyze large amounts of data while maintaining granularity of analysis
  • Tools to optimize and accelerate model risk management
  • Reporting requirements within financial services 
  • Pre-trade risk controls and best execution analysis

Mohammad Yousuf Hussain, Senior Technology & Innovation Specialist, Applied Innovation & Strategic Investments, HSBC
Aditya Mehta, Innovation Analyst at Applied Innovation and Strategic Investments, HSBC

17:30

End of Day 1

Day 2 - Investments

08:30

Registration 

09:00

Machine learning and trade strategies 

  • Finding alpha - value investing 
  • Factor investment 
  • Reinforcement learning 
  • Q learning 
  • AI for ESG 
  • Sentiment analysis 

Harsh Prasad, Vice President, MORGAN STANLEY 

10:30

Morning coffee break

11:00

Machine Learning and Portfolio Construction in Fixed Income - Part I 

  • Current status of the use of machine learning algorithms in FI portfolios 
    • how ML is used in the equity world
    • the challenges of using ML in FI
  • Current evolution of ML algos 
  • Building fixed income portfolios: 
    • Problem definition – what is the goal, what are the constraints
    • Optimisation techniques
    • Use of machine learning algorithms: 
      • data – predict, clean, fix
      • dimension reduction
      • attribute and behaviour prediction (i.e. default)
      • allocation of assets in portfolio 
        • covariance prediction
        • index construction
  • Live example of building a portfolio (e.g. to replicate an ETF) 

Hillel Raz, Chief Scientist, BondIT

12:00

Lunch

13:00

Machine Learning and Portfolio Construction in Fixed Income - Part II 

  • Building fixed income portfolios: 
    • Problem definition – what is the goal, what are the constraints
    • Optimisation techniques
    • Use of machine learning algorithms: 
      • data – predict, clean, fix
      • dimension reduction
      • attribute and behaviour prediction (i.e. default)
      • allocation of assets in portfolio 
        • covariance prediction
        • index construction
  • Live example of building a portfolio (e.g. to replicate an ETF) 

Hillel Raz, Chief Scientist, BondIT

14:00

Afternoon coffee break 

14:30

Machine learning in finance: practice

  • ML for quantitative investment: challenges and opportunities
  • Training equity model: features engineering and domain knowledge 
  • Non-linear multi-factor signal
  • Stock selection using gradient boosting model

16:00

End of Day 2