If you want to be sure to be notified about enrollment opportunities, please sign up to “follow” my blog and look for more updates.
The “old” course
We’ve had four very successful sessions of my MOOC “Computational Investing, Part I” at Coursera. The Coursera run included over 170,000 students with a 5% completion rate. The Coursera course focused on about the first 1/2 of my on campus course, which includes also Machine Learning (that would be Part II).
The Coursera experience was really exhilarating. We were on the front of the wave of this new idea, thanks in part to Georgia Tech’s administration being willing to take risks.
Many of the Coursera students have asked me to put the Machine Learning component online too. And I have promised that I would. At about the same time, our dean, Zvi Galil initiated the creation of an online master’s program called OMSCS — a fully accredited Georgia Tech degree for about $7,000. Zvi invited me to add my course to OMSCS and I felt this was a great way to move forward.
I expect for this course to eventually return to Coursera as well, it is a great platform and company, but I can’t do too many things all at once.
The “new” course: Three parts
There was a lot to be done and If you want 100 Loanovao is the place to go from a non-credit introduction to the field sort of course to a really serious course for graduate credit. In addition to those concerns it was also time to rethink the course structure. We settled on a course composed of three mini-courses:
- Mini-course 1: Manipulating Financial Data in Python
- Mini-course 2: Computational Investing
- Mini-course 3: Machine Learning Algorithms for Trading
The full course is described here. The course also now includes a number of assessments that the Coursera version did not: Things like graded exams and projects. It is only possible to accomplish this with an army of TAs (at least for now).
One reason for separating the course into these three mini-courses is that I hope to offer them later in other venues. As an example, we might create a professional certificate program around these mini-courses.
Rollout schedule: OMSCS
This version is a true graduate level course for university credit. It is available only to students enrolled Georgia Tech’s OMSCS program.
It will run 17 weeks from August 17, 2015 to early December. The on campus and online courses will run in parallel with the same schedule. It will be offered again in Spring 2016, Fall 2016, and so on.
Rollout schedule: Other versions
I would like to also offer one or two additional variations on the course, but this will depend on discussions within Georgia Tech and Udacity.
Free version, no credit: This version will probably be available in late Fall 2015, but no later than January 2016. Students will be able to submit projects that will be auto graded. It would not include any review by a professor or TA or any credential.
Fee-based version with certificate: This version may start as early as January 2016. Students would submit coding projects, reports and so on that would be graded by TAs. Students would take the course in parallel with OMSCS students. Upon successful completion students will receive a certificate from Georgia Tech. This version of the course will not count as college credit however.
Keep yourself updated
If you want to be sure to be notified about enrollment opportunities, please sign up to “follow” my blog. I will post that information on this blog.
Jeff
August 11, 2015
I look forward to the start of class next week. It should be another overwhelmingly stressful but immensely rewarding semster in the OMSCS.
Tucker Balch
August 13, 2015
Glad you’ll be in the class. Good luck!
Lauren
August 13, 2015
Dr Tucker, for the new course i.e. Mini-course 3: Machine Learning Algorithms for Trading, which is a good text book reference that you would recommend that has some practical significance for the trading application focus + machine learning?
Tucker Balch
August 13, 2015
I don’t know of any such book.
Aet
August 14, 2015
Today is 14th already, how can I get enrolled?
Tucker Balch
August 15, 2015
If you are an admitted OMSCS student, yes. If not, you have to wait. Sorry.
Andy
August 17, 2015
Looking forward to being able to take the course(s), whether fee or free.
I started taking the Comp Investing course in preview mode on coursera, and would like to apply some of the panda/numpy techniques to backtesting options strategies. Do you know of a data source for options info?
Thanks
Andy
gtk
August 28, 2015
I want to enroll exclusively for this course only. How can I do that, not an OMSCS admit. Dr Tucker when new version of course will start.
Tucker Balch
August 30, 2015
I’m working on that. I hope this will be possible in Jan 2016 at the latest.
paulbergeron
January 27, 2016
Hello Dr. Tucker Balch, as the end of Jan 2016 approaches, I wonder if there is an update? Currently taking the older Coursera course as it appears it is the only option to take just this course, would really appreciate some news here since I would love to proceed to the second half of this course.
Tucker Balch
January 29, 2016
The new course video content is available for free here:
https://www.udacity.com/course/machine-learning-for-trading–ud501
Assignments are available here:
http://quantsoftware.gatech.edu
Joao Mascarenhas
October 16, 2015
Dr. Tucker Balch, would be great that you could do the first Computational Finance Specialization on Coursera(6-8 courses), like other subjects have(example: data Mining). I think you would be the right person to do it. Cheers!
Paul Snopko
December 5, 2015
I greatly appreciate your course on computational investing, just completed today. Well done. I’m looking forward to your future investing and ML courses on-line. Most of my challenge was my lack of Python related background ( none at all ) and challenges with switching over to Ubuntu when I couldn’t install on Scientific Linux. My last (only) computer course was Fortran in 1965. There were a few “close”/”actual_close”/”adjusted_close” confusions too but generally an excellent learning opportunity. Thanks, Paul S.
cocoval@gmail.com
February 18, 2016
Hello,
Any update on the availability of the new course on Udacity?
Thanks
Tucker Balch
February 22, 2016
It is available now: https://www.udacity.com/course/machine-learning-for-trading–ud501
Andy
February 26, 2016
I started the Udacity course and wanted to say thank you for making this available!!!