• C++ Programming for Financial Engineering
    Highly recommended by thousands of MFE students. Covers essential C++ topics with applications to financial engineering. Learn more Join!
    Python for Finance with Intro to Data Science
    Gain practical understanding of Python to read, understand, and write professional Python code for your first day on the job. Learn more Join!
    An Intuition-Based Options Primer for FE
    Ideal for entry level positions interviews and graduate studies, specializing in options trading arbitrage and options valuation models. Learn more Join!

GRE Math Subject Test

Joined
8/15/11
Messages
62
Points
18
Would a good score on the GRE Math Subject Test provide a boost to an MFE applicant with an undergraduate degree in accounting? If so, what type of score would admissions folks consider "good"? Thanks.
 
Would a good score on the GRE Math Subject Test provide a boost to an MFE applicant with an undergraduate degree in accounting? If so, what type of score would admissions folks consider "good"? Thanks.

I would like to add a follow up question, can a Math GRE replace a general GRE test?
 
Would a good score on the GRE Math Subject Test provide a boost to an MFE applicant with an undergraduate degree in accounting? If so, what type of score would admissions folks consider "good"? Thanks.
75% percentile or above is considered good.
I would like to add a follow up question, can a Math GRE replace a general GRE test?
No. General GRE is required by all programs while Math GRE is required by one or two programs under specific situation.

It should be obvious that Math GRE test is much much harder and you are more likely to score lower in it than getting the max GRE Q score.
 
Thanks Andy. How beneficial is a score 75% + paired with non-degree coursework to fulfill prerequisites to an applicant lacking an undergraduate degree in a quantitative field?
 
Your question is essentially "how do I stand out from hundreds other highly qualified applicants who has Math/CS/Engineering degree"?
The answer is certainly not going toe to toe with them, specially in a short-cut way like taking the GRE Math.
Play to your strength, tell your story as a "finance" applicant, not as "CS/Math" guy. Every program wants to diversify their intake pool and as long as you meet the requirements and show you can perform, you have a good shot.
 
The Math GRE Subject exam is mainly of interest to Math PhD programs, who need it to distinguish among their applicants, who should have all gotten 800Q on the (now retired) GRE general exam, which doesn't differentiate sufficiently among those on the far right tail of the curve.
 
I don't know your background, but the GRE Math Subject Test is hard if you are a math major. I think if you took it, it would just be a waste of your money.
Do you know Abstract Alg, Linear Alg, Real Analysis, Complex Analysis, Discrete, Calc based Stats, Topology, Calc 1-3,DE, Number Theory, and Set Theory? You need to know all major theorems like the back of your hand and have computational ability in all fields.

I would like to add if you don't have a solid math background, your score will probably be a 200. That means you didn't answer one question. However, every question you get wrong is a negative 1/4. Google GRE Math Subject Test and download the practice book. I can tell you from experience that the practice book is 1000x easier than the test.
 
I don't know your background, but the GRE Math Subject Test is hard if you are a math major. I think if you took it, it would just be a waste of your money.
Do you know Abstract Alg, Linear Alg, Real Analysis, Complex Analysis, Discrete, Calc based Stats, Topology, Calc 1-3,DE, Number Theory, and Set Theory? You need to know all major theorems like the back of your hand and have computational ability in all fields.

The questions in most of the areas -- set theory, group theory, topology, number theory, linear algebra -- are pretty basic. For example, in group theory if you know the basics like Lagrange's theorem, cosets, ideas about normal subgroups, quotients, symmetries, etc., you should be fine. You do need a solid background in Calc 1-3, and speed in solving non-trivial Calc problems: the GRE math subject test is a glorified IQ test for math majors -- which means either you have the raw speed and ingenuity wired into you, or you don't.

The test is probably irrelevant for a prospective quant student.
 
I don't know if you have taken the test recently, but I can tell you from the one I took, Nov 2010, I needed to know almost every major Theorem in all subjects. My test wasn't full of that many computations. Knowing how to use Lagrange to compute |G| or |<g>| wasn't asked because that is simply divisions. The computations I had on my test were Calc based but there were less than 20 problems like that.
 
the GRE math subject test is a glorified IQ test for math majors -- which means either you have the raw speed and ingenuity wired into you, or you don't.

Not to veer too far off topic, but I disagree. I went through a self-designed training program and improved my percentile from 44th to 88th, and I know at least one other guy who did the same. I know other applicants whose "IQs" were at least as good as mine (whatever that means) who didn't perform as well.

That said, having a high (math) IQ may be a sufficient condition for scoring well; I know one such fellow who scored in the 90s (%) without putting in any effort.

The test is probably irrelevant for a prospective quant student.

Agreed.
 
That said, having a high (math) IQ may be a sufficient condition for scoring well; I know one such fellow who scored in the 90s (%) without putting in any effort.

It may not be sufficient to score one's best: everyone's performance goes up if they've been through some training program and mock exams. Which is what I was assuming. The 88th percentile may reflect your true standing -- i.e., regardless of how much extra effort you put in, you might not reach the 95th percentile. The person who scored in the 90s might have scored a few points higher with a bit of training. Same holds for the regular IQ test.
 
everyone's performance goes up if they've been through some training program and mock exams. Which is what I was assuming.

We're on the same page, then. I just wanted to clarify this point for prospective test-takers who might be reading this.
 
We're on the same page, then. I just wanted to clarify this point for prospective test-takers who might be reading this.

Well then, for those who are reading this, what -- *delicate cough* -- were the secrets of your training program? Did you overlearn the top ten or fifteen theorems in each area and aim for computational speed in mock exams? For example, in complex analysis, did you overlearn things like Cauchy's theorem, Residue theorem, Rouche's theorem, integration over paths, Morera's theorem, Cauchy-Riemann equations, etc.?
 
Well then, for those who are reading this, what -- *delicate cough* -- were the secrets of your training program? Did you overlearn the top ten or fifteen theorems in each area and aim for computational speed in mock exams? For example, in complex analysis, did you overlearn things like Cauchy's theorem, Residue theorem, Rouche's theorem, integration over paths, Morera's theorem, Cauchy-Riemann equations, etc.?

Extreme pragmatism. There are four real practice tests available. Take all of them under timed conditions (for the experience of pressure), then go over ALL questions, including those omitted. This gives you a practical feel for the types of questions they ask. Then simply design your own timed "tests" based on those kinds of questions--both in your strengths and weaknesses.

My "tests" included material from:

1) Calculus problems: pretty much all the basics, but no further than Green's theorem in multivariable. I overlearned formulas such as variations on sum(0..inf or 1..inf) of power series 1/(1-x) and 1/(1-x)^2. Also, Leibniz' Rule for differentiating an integral function wrt a variable that is an upper or lower bound of integration (or both) is VERY prominent on the exam. (As in, you're almost guaranteed to get such a problem on every exam.)

2) Complex Analysis (basic elementary functions, basic generating functions, Demoivre--very important, Cauchy's thm, Laurent series, basic residue theory, CR eqns, *not* Rouche)

3) precalc problems (probably all topics; don't remember exactly, though I hear this past year conics were emphasized)

4) linear algebra (basic problems in dimension, rank, linear independence, 2x2 inversion, eigenstuffs, basic vector spaces)

5) Algebra: basic problems in group, ring, and field theory (using Gallian :p ; mostly group stuff)

6) Elementary number theory (mostly solution sets of linear congruence equations; a little Chinese Remainder theorem)

7) Combinatorics: mostly just counting problems from the "Eightfold Path." During the actual exam I solved one such problem in my head during a relaxing bathroom break. Basic graph theory can come in handy (no further topics than what you would be expected to learn during a general discrete math course.) Also, generating functions and recurrence relations are important, and I'm pretty sure I practiced them.

8) Differential equations: mostly linear ODEs and basic exact equations. Homogeneity comes up here, in linear algebra, and ring questions (for example)

That's the bulk of it. Basic knowledge of topology and lebesgue theory could help you knock out some easy ones, but only if you know them ahead of time (i.e. don't learn these only for the exam.) Basic real analysis can also help you knock out some of the Roman-numeraled "I-IV" questions--topics seen on tests include MVT, basic counterexamples...and some common sense.

Now, I have heard the test has gotten a little harder since I last took it, in April 2009, so any prospective taker should do further research, especially in the Math gre forum.

Other sources:
http://www.mathematicsgre.com/ (Math GRE forum)
http://www.math.ucsb.edu/mathclub/GRE/ (Three old practice exams; fourth available via ets, maybe)
Sorry if this was a bit rushed. I'm in the middle packing to move to my new graduate school :eek:
 
The practices exams don't do the test any justice now. I took the test in Nov. 2010, and the taking the practice test didn't help. Even the practice test supplied by the GRE isn't anyway comparable to the test they will give you on test. However, there are 3 versions. Version C is the hardest and you need to get less right to get a higher score. For example, a tester taking A may need 44 right for a 700 whereas someone taking C will need 37.
 
The practices exams don't do the test any justice now. I took the test in Nov. 2010, and the taking the practice test didn't help. Even the practice test supplied by the GRE isn't anyway comparable to the test they will give you on test. However, there are 3 versions. Version C is the hardest and you need to get less right to get a higher score. For example, a tester taking A may need 44 right for a 700 whereas someone taking C will need 37.

That's too bad. In that case, I'd conjecture that the best way to score well (on top of the aforementioned training) is to simply be good at higher mathematics topics: Analysis, Topology, Algebra, etc. If you know the basic theorems, counterexamples, and proof forms of those, you can knock out some of those questions without too much difficulty. There is, of course, no shortcut to learning those.
 
Back
Top