Course Overview

This 6-lecture (9 hours) course provides an overview of the application of statistical and machine learning (supervised learning, unsupervised learning) in the financial industry. We will provide an introduction of career options and paths on relevant jobs, dive deep into different kind of directions of the industrial applications by going through the technical aspects. Theses directions include credit risk, fraud strategies, breakage modelling, campaign analytics, credit strategies, recommendation systems, forecasting, and anti-money laundry. After the lectures, students will implement a project with self-selected topics, with the lecturer’s offline 1-1 supervision.

Lecture 1 – Introduction to Credit Risk Modelling

– Definition of credit risk, and the framework of credit risk modelling

– Various kinds of credit risk models, such as PD, LGD, EAD, PPNR, etc.

– examples of model selection and feature selection

– Credit risk modelling lifecycle in financial institution

Lecture 2 – Credit risk modelling deep dive

Common tests in the industrial practice of credit risk model development and model validation, such as 

– assumption checking 

– back testing

– conservatism assessment

– scenario analysis

– benchmarking

– sensitivity analysis

– model monitoring

Lecture 3 – Other Modelling Topics (1)

Other applications of supervised learning/predictive modelling and unsupervised learning in finance, such as breakage modelling, recommendation systems, economic forecasting, credit strategies, campaign analytics, and fraud strategies

Lecture 4 – Other Modelling Topics (2)

Other applications of statistical and machine learning models in the financial industry

– Economic Forecasting

– Credit Strategies

Lecture 5 – Other Modelling Topics (3)

Other applications of statistical and machine learning models in the financial industry

– Campaign analytics

– Fraud Strategies

Lecture 6 -Typical questions on statistical models and machine learning in interviews of relevant positions, such as

– Credit risk modelling quantitative analyst

– Model validation quantitative analyst

– Customer analytics data scientist

– Fraud strategy analyst

– SQL questions with typical difficulties

– Business knowledge in the line of practice

– etc.

Course Highlights

Establish an overview of statistical and machine learning modelling positions in the financial industry. Understand the type of such positions, their responsibilities, job requirements, and connections to academic formal trainings. Learning in-depth how the models are used in the industry in various applications, gaining hands-on experience on a project which aligns the industrial practice, and eventually get prepared for job seeking on this career path from various aspects.

Qualifications

This course is suitable for students with some training on statistical models (e.g. regression, time series)/machine learning/predictive modelling on undergraduate level and above, with some exposure to a computer programming language (e.g. Python, R, SAS, Matlab).

Mentor Profile

The mentor for this course is currently a Senior Associate in the Financial Engineering Team of Deloitte Canada. In Deloitte, he is primarily working on quantitative modelling practices in financial engineering and risk management. Prior to joining Deloitte, he had a data science position in RBC which was primarily responsible to oversee the patterns and trends in the banking book related to fraud activities. Mentor has a throughout understanding on machine learning and statistical models, especially their applications in the financial industry, such as risk modelling, asset pricing, portfolio optimization and trading, and commercial banking analytics (business strategy, campaign, fraud, AML, etc.) He has a Master’s degree in Operations Research from the University of Toronto, where he focused on research in Deep Reinforcement Learning, and its application to trading.