Fintech Series Part 4: How Are We Addressing the Challenge?

Welcome to the final part of my Fintech Series. Be sure you’ve checked out part one, part two, and part three.

Developments in fintech pose a new challenge to traditional finance curricula. As discussed in part three, the quant needs a more technical background to succeed in finance. We, at the Berkeley MFE Program, are addressing the challenge by preparing our students with thorough pre-program coursework, a curriculum that emphasizes Python, R, machine learning, and artificial intelligence, industry projects with both long-established financial institutions and new tech companies, and through the roll-out of a second cohort with an emphasis in data science. 

The Berkeley MFE Program has always been the leader when it comes to preparing students. We were the first ones to launch pre-program courses in 2002; pre-program coursework in statistics, in mathematics (PDEs and numerical analysis) and C++ which used to be the industry programming standard back then. We continue to be the only ones who work one-on-one with each student prior to the start of our program to ensure his or her success at Berkeley and beyond. While every school talks about data analytics, we have quietly prepared our students for the increasing demand for specific data science skills since 2012.

Our students are well versed in Python, R, machine learning, and artificial intelligence as we integrate all of these concepts in the curriculum. We even provide industry projects for our "pre-students"- I call them "pre-students" because while the MFE Program starts in March, the students already work remotely with industry professionals from the buy-side and sell-side and other firms on key projects prior to the official start of the program. Many jobs from both the sell-side and buy-side demand machine learning and data science skillsets, thus we have focused on making our students the most prepared and desirable candidates for the financial and tech industries. We have placed some alumni and current students at firms that require a strong background in data science.

Berkeley MFEs have a wide variety of skills: strong knowledge of statistics, mathematics, programming, finance, economics, and great logic and intuition. Taking a data analytics certificate or boot camp series of a few courses is not sufficient to make you a great data scientist if you lack financial intuition - something all of the MFE courses focus on. I believe we have nailed down the best curriculum because we prepare our students and make them work on industry projects from the get-go. I remember presenting our program to a Chief Data Scientist at a top tech firm, and after hiring two of our alumni, he was determined to hire more Berkeley MFEs.

This term, current students have the opportunity to take courses with emphasis on data analytics. In the "Asset-Backed Securities" course, Dr. Nancy Wallace added a big data lecture that includes concerns about statistical discrimination in credit scoring; there are homework sets on new loan-level data. The homework is more closely tied to Bloomberg accessible data and industry practice for the term structure modeling. Additional topics on esoterics and catastrophe risk analytics will also be covered.

The “Building Machine Learning Systems: Tools, Platforms, and Financial Applications” course this term is an introduction to building machine learning systems for data science in industry, illustrated with financial and other applications.  Taught by Carolina Galleguillos and Gert Lanckriet, it covers topics ranging from working as a data scientist, formulating data science problems, and handling real-life data, to building machine learning systems, and communicating insights. It will highlight the most popular tools and platforms used today, and overview machine learning algorithms for different classes of problems, with case studies from industry. Students will gain hands-on experience formalizing machine learning problems, collecting, analyzing and processing real data, turning statistical and machine learning concepts into practical machine learning solutions and launching those in production.

Carolina Galleguillos is a machine learning engineer at Thumbtack, where she designs and develops machine learning solutions for customer growth and user personalization. In 2017, she was named on Forbes’ list of 20 Incredible Women Advancing A.I. Research.

Gert Lanckriet leads the Machine Learning team at Amazon Music, and is a Professor of Electrical and Computer Engineering at UC San Diego. His interests are in data science, on the interplay between machine learning, applied statistics, and large-scale optimization, with applications to music and video search and recommendation, multimedia, and personalized, mobile health.

Effective March 2018, there will be two cohorts of MFE students - 40 students each - for a total of 80 students. 

What does this mean? The degree title remains the same, and the curriculum too for the most part. So why two cohorts? One will have an emphasis on data analytics, or data science, and students will select their electives and work with fellow students based on their field of interest.

Does it mean that all courses will be taught twice? No, as it will depend on the number of students in the elective courses, and the professors’ availability and choice. Most of our core courses will have two sections – i.e. for cohort 1 and cohort 2 - with the exception of the Applied Finance Project course. Both cohorts will start at the same time - at the end of March 2018. Orientation will be held for all 80 students together in mid-March 2018.

Indeed these are exciting times for the MFE!  Interested in joining us? The application deadline for our next class (starting in March 2018) is October 2, 2017. Join us for an info session or start your application now.

For the most up-to-date info, follow us on Facebook, Twitter, and LinkedIn.  Learn more about the Berkeley MFE Program and how we can help you launch your career by going to our website.