How many courses are appropriate for the MFE degree in one semester?

[Terry.T]

Hello everyone. I am about to join the MFE program at Columbia University this fall. I am currently thinking about choosing several courses in a semester. I know that most of the MFE programs are four courses in a semester, but Columbia allows students to complete the program in a year with 6 courses in a semester. I don't know if 5-6 courses in a semester are feasible. Will I have no time to invest in finding a job because of the stress of the course? Thank you all!

 

[noether-skolem]

Drop me a pm with your columbia email, I can introduce you to someone who did 6 courses/semester last Fall and this Spring.

 

[jarryds]

I was wondering the same thing. Looks like the institutions recommend four courses, but five should not be too difficult? My plan is to take four the first semester and add on for future semesters.

 

[Michsund]

Same thing is recommended at cmu. Take the base classes and if you do good add more if wanted

 

[noether-skolem]

If the goal is to finish the program in two semesters (Fall + Spring) you'd have to start at full speed and take 6 courses in both Fall and Spring terms though. Definitely doable but how stressful it'll be and how much time you'd have left for job hunting would obviously depend on the person.

 

[Terry.T]

If the goal is to finish the program in two semesters (Fall + Spring) you'd have to start at full speed and take 6 courses in both Fall and Spring terms though. Definitely doable but how stressful it'll be and how much time you'd have left for job hunting would obviously depend on the person.

If the six courses in a semester are all courses that require a lot of time investment, then it will definitely squeeze the time to find a job. I think this is why Columbia MFE students choose to graduate in one and a half years. Probably choosing 6 courses means choosing one or two very relaxing courses to balance the time.

 

[jeffsnguyen]

I suggest factoring in time for recruiting (including interview prep) & networking. Those are very time consuming tasks, especially if your school career center isn't actively looking out for you like a few programs. I personally would prioritize them over trying to overloading as many courses as possible. As with many things in life, it's a marathon, not a sprint. Runners will tell you start easy so you can finish strong.

 

[noether-skolem]

If the six courses in a semester are all courses that require a lot of time investment, then it will definitely squeeze the time to find a job. I think this is why Columbia MFE students choose to graduate in one and a half years. Probably choosing 6 courses means choosing one or two very relaxing courses to balance the time.

Before you dive into 6 courses/semester, make sure you're damn good at stochastic and coding. Even the "relaxing" courses may not be as relaxing as you expect, if you don't have very solid foundations in maths/programming, since we're only allowed to work with a curated list of courses to satisfy the degree requirement (you can't take some random easy courses that's unrelated to the MSFE curriculum, those won't count towards the 36 credits needed to graduate).

That said, it is absolutely possible. The person I know did 6 courses in Fall and 6 courses in Spring, ended up with a good GPA, and landed a full-time job. If you're really confident about yourself, just go ahead and register for 6 courses, gauge the workload in the first few weeks and drop courses if you feel like you're not up to it. I tried registering for 5 courses in Spring and decided to drop one, since all 5 of those were pretty heavy with the weekly/bi-weekly homework/assignments and projects.

 

[Andy Zhang]

If the six courses in a semester are all courses that require a lot of time investment, then it will definitely squeeze the time to find a job. I think this is why Columbia MFE students choose to graduate in one and a half years. Probably choosing 6 courses means choosing one or two very relaxing courses to balance the time.

Statistical Methods in Finance, which @Qui-Gon asked about a while ago, from the Stats department is pretty relaxing (and an easy GPA booster course) lol. Not sure if it can be counted though since there is a mandatory Statistical Analysis and Time Series course.

make sure you're damn good at stochastic and coding

I wanted to highlight this part because it is so f***ing true.

For stochastic, unless you've had rigorous treatment of it, e.g., you took graduate/undergrad level stochastics at a school like UC Berkeley, Dartmouth, etc, you will likely struggle (ofc, you could be an anomaly from the rest of the crowd). If I recall correctly (and @noether-skolem can correct me on this), the midterm and final exam weight distribution are equal. So if you f*ck up the midterm, there'll be a lot of pressure to do well on the final. Also, I'd try spending time to really understand stochastics since it's super helpful on interviews.

For coding, unless you have a strong computer science fundamentals background paired with coding interview experience, I would strongly recommend spending time doing leetcode so you can get used the type of coding questions you'll encounter. In interviews, you might get asked to either solve a dynamic programming (bolded cuz it's usually the hardest part of a coding assessment) question (can be quite hard even if you have the appropriate background from a top school) in Hackerrank or some other algorithm question on a white board. If you never done this before or only did it a few times, it can be quite daunting with the added time pressure.

Anyways, if you're fine with stochastic and coding, then go for it!

 

[Terry.T]

Before you dive into 6 courses/semester, make sure you're damn good at stochastic and coding. Even the "relaxing" courses may not be as relaxing as you expect, if you don't have very solid foundations in maths/programming, since we're only allowed to work with a curated list of courses to satisfy the degree requirement (you can't take some random easy courses that's unrelated to the MSFE curriculum, those won't count towards the 36 credits needed to graduate).

That said, it is absolutely possible. The person I know did 6 courses in Fall and 6 courses in Spring, ended up with a good GPA, and landed a full-time job. If you're really confident about yourself, just go ahead and register for 6 courses, gauge the workload in the first few weeks and drop courses if you feel like you're not up to it. I tried registering for 5 courses in Spring and decided to drop one, since all 5 of those were pretty heavy with the weekly/bi-weekly homework/assignments and projects.

Very wonderful answer. Thank you for your opinion on this issue!

 

[Terry.T]

Statistical Methods in Finance, which @Qui-Gon asked about a while ago, from the Stats department is pretty relaxing (and an easy GPA booster course) lol. Not sure if it can be counted though since there is a mandatory Statistical Analysis and Time Series course.

I wanted to highlight this part because it is so f***ing true.

For stochastic, unless you've had rigorous treatment of it, e.g., you took graduate/undergrad level stochastics at a school like UC Berkeley, Dartmouth, etc, you will likely struggle (ofc, you could be an anomaly from the rest of the crowd). If I recall correctly (and @noether-skolem can correct me on this), the midterm and final exam weight distribution are equal. So if you f*ck up the midterm, there'll be a lot of pressure to do well on the final. Also, I'd try spending time to really understand stochastics since it's super helpful on interviews.

For coding, unless you have a strong computer science fundamentals background paired with coding interview experience, I would strongly recommend spending time doing leetcode so you can get used the type of coding questions you'll encounter. In interviews, you might get asked to either solve a dynamic programming (bolded cuz it's usually the hardest part of a coding assessment) question (can be quite hard even if you have the appropriate background from a top school) in Hackerrank or some other algorithm question on a white board. If you never done this before or only did it a few times, it can be quite daunting with the added time pressure.

Anyways, if you're fine with stochastic and coding, then go for it!

Thanks for your suggestions!

 

[Daniel Duffy]

5

 

[Qui-Gon]

I’d venture to say that maybe the top 10% of students could reasonably handle 5+ courses in the top programs. I would say for everyone else (and you have to be honest with yourself if this is where you lie), stick to 4 and learn the material thoroughly. The average student doing 5+ technical classes is likely predominantly worried about simply getting assignments finished and will have little time to properly consume the material which should be a student’s main priority.

If you really want to take more than 4 courses, my advice is the following. Pick 2 of the classes (ones closest to your area of comfort) and get your hands on their syllabi and past homework assignments. Read the textbooks used and work through the assigned problems (don’t have to do all of them, but enough to get a feel for the style of questions and how to approach solving them) over the winter/summer break prior to the semester. If done honestly, this is a useful strategy for being able to contend with 5+ courses in one term.

 

[noether-skolem]

For stochastic, unless you've had rigorous treatment of it, e.g., you took graduate/undergrad level stochastics at a school like UC Berkeley, Dartmouth, etc, you will likely struggle (ofc, you could be an anomaly from the rest of the crowd). If I recall correctly (and @noether-skolem can correct me on this), the midterm and final exam weight distribution are equal. So if you f*ck up the midterm, there'll be a lot of pressure to do well on the final. Also, I'd try spending time to really understand stochastics since it's super helpful on interviews.

Yeah for most courses it's like 70% of the final grade being roughly split between midterm and final exam (could be same, could be slightly tilted towards final - but never the other way). Exams are always high pressure, remember that the top programs on the QN ranking admit the best applicants from all around the world, even the "below average" student in Princeton/CMU/Columbia/Baruch could've been a top-tier student in a less selective program.

You also nailed the part about coding interviews, dynamic programming can be very tricky in a phone interview, since on one hand you don't want to always say "ummm" but on the other hand you also don't want to say something completely irrelevant - there is significant pressure in quickly coming up with *some* ideas (even if they don't work, if they're somewhat in the right direction, you can elaborate and expand on it). Solving a coding assignment is very different from coming up with and explaining your algorithm on the fly.

 

[noether-skolem]

I’d venture to say that maybe the top 10% of students could reasonably handle 5+ courses in the top programs. I would say for everyone else (and you have to be honest with yourself if this is where you lie), stick to 4 and learn the material thoroughly. The average student doing 5+ technical classes is likely predominantly worried about simply getting assignments finished and will have little time to properly consume the material which should be a student’s main priority.

If you really want to take more than 4 courses, my advice is the following. Pick 2 of the classes (ones closest to your area of comfort) and get your hands on their syllabi and past homework assignments. Read the textbooks used and work through the assigned problems (don’t have to do all of them, but enough to get a feel for the style of questions and how to approach solving them) over the winter/summer break prior to the semester. If done honestly, this is a useful strategy for being able to contend with 5+ courses in one term.

Yeah preparations before starting the program/semester definitely helps. In hindsight I probably could've handled 5 courses in my first Fall semester, but on the other hand I might've had to spend less time on job hunting and interview prep (well, I did end up getting a good internship at a pretty established buy-side firm so I'm not complaining).