Machine Learning and AI Techniques

The popularity of data science techniques such as data mining and machine learning has grown enormously in recent years. They present effective solutions to process and analyze the huge amount of data available to risk managers and financial analysts.

With the advances in computing power and distributed processing, it is now possible to process - and make sense of - the vast array of information that can be gathered from several different data sources.

This hands-on program covers key techniques - including several aspects of supervised and unsupervised machine learning - that can be used when mining financial data. The program also focuses on advanced data science techniques that are becoming widely used in financial markets for text analysis and Artificial Intelligence (AI): Natural Language Processing (NLP) and Deep Learning (DL).

The program is delivered entirely through workshops and case studies. Participants will learn how to implement natural language processing techniques by building a sentiment analysis model to analyze text. In the deep learning section, participants will focus on the different neural networks that can be put at work for data classification, time-series forecasting, and pattern recognition.

All exercises and case studies are illustrated in Python, allowing you to learn how to work with this flexible, open-source programming language.

Basic programming experience in Python is recommended, which can be acquired in the 2-day LFS Python for Finance program.

This course is primarily aimed at those working in financial institutions; as well as regulatory bodies, advisory firms, and technology vendors. Specific job titles may include but are not limited to:

• Trading
• Portfolio management
• Asset allocation
• Data science
• Financial engineering
• Quantitative analytics and modeling
• Infrastructure and technology

• Build a solid knowledge base on data mining techniques and tools, as well as their application to the financial industry
• Gain hands-on experience with Natural Language Processing and Deep Learning in finance
• Learn how to apply Python to data mining and processing, and to solve real-world NLP and DL problems
• Gain an understanding of Artificial Neural Networks (ANN) algorithms and how to use them to design, build, and develop DL models

• Basic notions of statistics
• Good working knowledge of Excel
• Knowledge of Python is required

Positioning of Machine Learning vs. Deep Learning

Machine Learning Introduction

• Supervised vs. unsupervised
• Association rules
• Classification vs. regression problems
• Cross validation and hyper parameter optimization

Unsupervised Learning

• Clustering analysis

Workshop: Equity / credit models

• Outlier detection
• Distance Metrics in Sklearn

Workshop: Robust outlier detection

• Kernel Density Estimation

Workshop: BitCoin-application

• Hidden Markov Models

Workshop: GBPEUR-timeseries analysis

Supervised Learning

• Regression with regularisation
  • Ridge regression
  • Lasso
  • Elastic Net

Workshop: Portfolio hedging

• Miscellaneous Regression Techniques
  • Gaussian Process Regression (GPR)
  • Principal Component Regression (PCR)
  • Partial Least Squares (PLS)

Workshop: Volsurface smoothing

• Classification
  • Naive Bayes classification: A straightforward and powerful technique to classify data
  • Linear Discriminant Analysis (LDA)
  • Logistic Regression

Workshop: Classification trees

Natural Language Processing

• Extracting real value from social media posts, images, email, PDFs, and other sources of unstructured data is a big challenge for enterprises
• Explore and tokenize a text
• Sentiment analysis
• Text Classification
• Understanding concepts such as WordNet, Word2Vec, Stemming, etc.

Workshop: Sentiment analysis of tweets

Deep Learning (AI)

• Deep Learning as a subfield of machine learning - Artificial Neural Networks (ANN) algorithms
• Forward and backward propagation
• Network topology
• Tensorflow 2.0

Workshop: Regression, classification, and time series forecast