Applications of Machine Learning to Financial Data
Project Description
Machine learning (ML) is already an indispensable tool in research. This will be increasingly true in the future.
Through this round of UROP, we will learn how ML (and more traditional statistical tools) can be applied in fields of financial economics (e.g., banking, insurance, and asset pricing).
The above description is purposefully general, given that students will be matched to appropriate projects depending on their interests and skill levels. At the same time, all projects will involve the use of ML in financial economics.
* Important note: This project is geared towards students with plans for academic rather than industry careers. Priority will be given to students who hope to apply to graduate programs in economics or finance.
Through this round of UROP, we will learn how ML (and more traditional statistical tools) can be applied in fields of financial economics (e.g., banking, insurance, and asset pricing).
The above description is purposefully general, given that students will be matched to appropriate projects depending on their interests and skill levels. At the same time, all projects will involve the use of ML in financial economics.
* Important note: This project is geared towards students with plans for academic rather than industry careers. Priority will be given to students who hope to apply to graduate programs in economics or finance.
Supervisor
NOH, Don
Quota
7
Course type
UROP1100
UROP2100
Applicant's Roles
The applicant will be expected to work on all stages of the project including, but not limited to, data collection (e.g., web scraping), data cleaning (e.g., Python + Pandas), and statistical analyses. The following is a non-exhaustive list of possible tasks:
(1) Process stock market and/or banking data, mainly with Python.
(2) Run regressions (time-series, Fama-MacBeth) and predictive algorithms (tree-based or linear).
(3) Make tables and figures that will be used for publication.
(1) Process stock market and/or banking data, mainly with Python.
(2) Run regressions (time-series, Fama-MacBeth) and predictive algorithms (tree-based or linear).
(3) Make tables and figures that will be used for publication.
Applicant's Learning Objectives
In increasing difficulty, the applicant will learn to:
(1) Learn the basics of economics and statistics.
(2) Learn to clean market data (e.g., dealing with missing data, outliers, fast computation with large data).
(3) Learn the basics of statistical analysis with Python.
(4) Learn to generate research ideas in financial economics and write papers.
(1) Learn the basics of economics and statistics.
(2) Learn to clean market data (e.g., dealing with missing data, outliers, fast computation with large data).
(3) Learn the basics of statistical analysis with Python.
(4) Learn to generate research ideas in financial economics and write papers.
Complexity of the project
Challenging