About the program:
Overall, my experience at UIUC was positive. The program was demanding, and although I joined some social organizations like the tennis club in my first semester, I was unable to find time later on to do anything aside from studying and looking for jobs. I learned a lot and gained a lot from attending the program. My technical/quantitative skills have vastly improved.
-The people at the MSFE office are very nice and accommodating, and I sense that they are truly committed to providing for their students and helping them succeed.
-The professors (for the most part) are very respectful of the students, and committed in helping them succeed.
-The student body, in my year, was too homogenous in culture. I think the program should strive to achieve a balance in class diversity – as they seem to have done for the 2015 class.
-The practicum is really what makes this program unique and exceptional. There is really no replacement for hands-on experience with an employer. You can think of the practicum as an apprenticeship. There was a lot of diffusion of knowledge between the sponsor and our team, and our sponsor was also nice enough to offer compensated summer work for students whom were unable to find summer internships. The practicum for us was where we really saw the interdisciplinary blend of practices come together in finance, as we saw how text-mining was used in finance for Twitter sentiment scoring, got to work with the data and build forecasting models, and also receive feedback on our work throughout the entire process. Skills obtained through the practicum could not be obtained through the classroom.
However, the biweekly meetings in Chicago was a bit strenuous for me particularly, because I also chose to take extra courses that were in session on Fridays. Going to Chicago takes about 3-4 hours one way by Greyhound, and we had to allocate every other Friday for the trip to go to Chicago for a 1-hour meeting with the sponsor. In the end, I was overloaded with work throughout the semester, and I wish I had the capacity to put in more time and effort for the project.
-There were too many mandatory meetings in the second semester in Chicago like the IAQF panel and the volatility workshop. I think mandatory meetings in Chicago should occur on Saturdays, so students taking classes on Fridays don't have to skip class.
-Career services/preparation really needs to ramp up.
About the University of Illinois:
Apart from the program, my experience at UIUC was very positive. I took 5 classes outside of the MSFE program (machine learning, finite element methods, partial diff. eqs, econometrics, and mathematical statistics), and they were all demanding, but truly world-class. This amount of coursework, combined with the rigorous contents, was a bit too much for me and affected my GPA, but I think the exposure I got was really worth it. It was an eye-opening experience where I got to take the same classes with statistics phds, economics phds, computer science phds, and mathematics phd students, and see for myself how I would do comparatively. The University of Illinois is truly a top university across all these fields and more. Financial engineering is by definition an interdisciplinary field, and I think it is important for any top institution to stay true to this interdisciplinary spirit, and offer elective coursework in all these areas. Given that the need for quant professionals changes constantly in the finance industry, I believe that a program that prescribes a fixed curriculum risks losing relevance to industry quickly. In this regard, the University of Illinois exceeded my expectations with the vast amount of top-notch coursework available. As a direct result of this, I was able to obtain interviews across a wide range of positions from algorithmic trader, to risk management consultant, to data scientist positions.
Apart from my review, I have a couple suggestions (ideas) on how the program can be improved:
1. Career Services (for Chinese Students)
- Make a separate career services specializing on getting Chinese students hired, and diffusing the sino-american cultural barriers. I think many of the Chinese students are technically talented enough to pass the technical interviews, but I would guess they must get held up in the behavioral interviews.
- I think that career services should already begin when the student enters the program, where students should identify a target position they want to apply to and receive counsel on how to strategize and build their resumes for those positions.
- Emily's occasional e-mails notifying us about open positions and events/tips weren't very helpful. I think quant interview workshops and role-playing workshops is what is really needed. Maybe bring in an outside quant interview expert for a day to hold a workshop or something(?).
- I would like to see some sign that the department is actively soliciting employers to come to campus and recruit MSFE students. At the present, the only visible relationship between the department and employers is the practicum, which I am hoping would change.
- Things they could use help on include building their resumes for specific positions through experiences and coursework, behavioral and technical interviews, finding the right jobs to apply to, how to approach recruiters at career fairs, understanding the typical recruitment process at American companies, understanding the power of acquainting with the recruiter beforehand, how to write follow-up e-mails, strategies on getting past the 'do you have work authorization' question, etc.
- Chinese students also need to recognize that they need to accept some American values and 'americanize' if they want to find jobs in U.S.
2. Re-evaluate the current curriculum
- While I realize that derivatives valuation is at the heart of the financial engineering discipline, the reality appears to reflect that the need for such people in industry is fast declining. Many of the tools acquired through an FE education are still relevant – like risk management, optimization, clustering, monte-carlo simulations, time-series modeling, and programming. However, it appears that current industry is more interested in using these tools within the context of data analytics. I am unsure of how relevant courses like financial economics and stochastic calculus are in the current industry, and they could be replaced to teach more data-oriented skills like Hadoop, cloud computing services like AWS, SQL, machine learning & data mining, data visualization, and statistical programming, that could make students more employable.
3. Standardize the programming language used throughout the curriculum, and then implement a baseline standard for programming skill across all students
- The employability of a student is highly correlated with that student’s programming skills. For a student to gain a high level of competency in a language, he/she needs to continue using it for a period of time. To make this process faster, I really believe that the program should just stick to one language used throughout the entire curriculum, and I further argue that the ideal language for this would be Python.
- With the current way things are, a lot of students graduate with little to mediocre programming skills, and most students graduate with Matlab being their primary language of choice. I believe this model is unsustainable because Matlab is rarely used in industry, and it is a superficially easy language to use. Programming skill in Matlab does not carry over to many other languages.
- Many students enter the program with no prior experience using Matlab, and then get introduced to C++ and R - all while still climbing the learning curve for Matlab. This results in them never fully climbing the learning curve for C++ or R or Python. Instead of trying to learn all these different languages at once, it's better to just concentrate on learning one main language that can be used in almost all situations, which isn't Matlab.
- Financial computing teaches us the type-safe language C++, which is useful to know, but a different beast from dynamic languages like Matlab, Python, and R. For the purposes of finding quant employment outside of quant development, knowledge of dynamic languages is more important, and Python seems to be the king in this field. Python seems to suite all the needs of a financial engineer in one language, and it is used widely in industry as opposed to Matlab.
- Python is probably the most frequently used language in the growing area of data analytics.
- I think that programming is important enough that the MSFE department should monitor and set a baseline standard for how well the students can do it.