Assessed Project
Submit online via QMplus before 23:59 London Time on Friday 28th April 2023
ECOM193 Statistical Machine Learning in Finance
This project carries a maximum of 100 marks. A 50% weighting will be applied to your overall
ECOM193 module score for completion.
You are required to submit a typed document in Word or PDF format containing written analysis
together with any supporting tables, graphics and code you think is necessary.
The details of what is required for this project are given on the next page.
This project has an maximum word count of 2500 words excluding computer output and any R
code that you choose to submit.
This project must be your own work presented in your own words.
Examiner: Dr R.A. Saldanha
© Queen Mary University of London, 2023
Page 2 Project Coursework: Semester B ECOM193 (2023)
Gradient Boosting
Boosting is a general ensemble method that aims to create a strong classifier from a number of
weak classifiers.
Explore and explain in detail the xgboost R package available at
https://cran.r-project.org/web/packages/xgboost/index.html. For example, what type of
problems is xgboost designed to solve? How exactly does it work? What is its parametrization?
Use xgboost to model the Credit Card Fraud Detection dataset (creditcard.csv) in an
appropriate manner of your choosing; see