September 2, 2020
Online Data Science in Finance course series from Purdue applies machine learning to modern financial problems
WEST LAFAYETTE, Ind. — Purdue University is offering a new series of Data Science in Finance courses focusing on applications of data science and machine learning to solve modern financial problems and leading to a certificate in the burgeoning field.
The three courses are 100% online, allowing flexibility for working professionals in the financial industry who want to advance their skills and careers, and making the course series readily available to an international audience of students. The courses also can be credited toward a new online master’s degree in Data Science in Finance that Purdue tentatively plans to begin offering in 2021.
The first course, Machine Learning in Finance 1: Learning the Fundamentals, will be offered starting Oct. 19. That course is an overview covering the basics of quantitative finance and machine learning, including probability and statistics and structuring financial data for use with machine learning algorithms.
The course also teaches basic Python programming, which will be used in the second and third courses focusing on practical applications of data science and machine learning in finance, as well as being a valuable professional skill generally.
Kiseop Lee, associate professor of statistics, described the initial course as “an appetizer sampler.” Lee and statistics professor Xiao Wang said the course covers a broad spectrum of knowledge that, in addition to being worthwhile on its own, lays the foundation for the second and third courses, to be offered beginning in Spring Semester 2021.
“If a student wants to get into this area, they need this fundamental knowledge,” said Wang, whose research focuses on artificial intelligence and machine learning while Lee’s focuses on quantitative finance. The two are co-developing and co-teaching the courses, bringing expertise in both aspects to the curriculum.
While some programs offered elsewhere combine elements of finance and data science, Purdue’s courses stand out by thoroughly integrating the two and heavily incorporating the machine learning element, Lee said. Using machine learning, professionals in the financial industry can employ statistical models to draw insights from their data to make strategic decisions, solve financial problems and increase profit.
Wang said the second course, Machine Learning in Finance 2: Reinforcement Learning, is broken into two halves. In the first half, students delve into artificial intelligence, neural networks and machine learning. Special emphasis is placed on “reinforcement learning,” in which software agents take independent actions within an environment representing a particular kind of problem, with the ultimate goal of maximizing rewards.
The second half of the course introduces modern finance problems involving topics such as hedging and pricing; portfolio dynamics and optimization; and high-frequency data. The curriculum then covers the use of machine learning, programmed in Python, to solve such problems.
The third course, Machine Learning in Finance 3: Application with Data Science, will focus on applying artificial intelligence methods in finance, particulary machine learning approaches for solving the most current challenges in financial markets. High-frequency markets and data-based trading strategies are topics of emphasis.
Each course is seven weeks and includes video lectures and online discussions, with quizzes and a final exam during the final week. Faculty members will provide assistance and answer questions during dedicated office hours through WebEx. The courses can be taken individually, but Machine Learning in Finance 1 must be taken first. Completing all the courses leads to a Data Science in Finance certificate from Purdue. Tuition is $1,800 per course.
For more information on Purdue’s Data Science in Finance certificate series and to register, go to https://science.purdue.edu/data-science/academics/online-finance.html.
Writer: Greg Kline, 765-494-8167, email@example.com
Sources: Kiseop Lee, firstname.lastname@example.org
Xiao Wang, email@example.com