EE or Math

[mhancock743]

Hello!

So I suppose in this thread I'm just asking for some general advice and insight. I guess I should start with some personal info:

Major: Math w/ minor in Comp Sci

GPA: 3.88 / 4

GRE: Q: 770 V: 530 A: 4.0

Graduation Date: Dec, 2011

Up until recently, I've been solid in thinking I would make the switch to EE by applying to various Master's programs in the Fall.

My interest would be signal processing. However, I've begun to think maybe of applying straight to a PhD program in Math instead

due to a newly sparked interest in Harmonic analysis, Fourier analysis and related topics (been combing through those free MIT lectures).

Some pros/cons:

I haven't yet taken the Math subject test since I planned on applying to EE. Nearly all require it for admission to Math programs

(surprisingly enough). The soonest available test date isn't until October, and I was hoping to have apps out by this time.

In applying to Math, there would be no 1-2 semester remediation period that I would expect with EE. On the other hand, EE also seems to offer a

more open career path.

Apologies for writing a novel, thanks for any feedback.

[StatlyDude]

Hello!

So I suppose in this thread I'm just asking for some general advice and insight. I guess I should start with some personal info:

Major: Math w/ minor in Comp Sci

GPA: 3.88 / 4

GRE: Q: 770 V: 530 A: 4.0

Graduation Date: Dec, 2011

Up until recently, I've been solid in thinking I would make the switch to EE by applying to various Master's programs in the Fall.

My interest would be signal processing. However, I've begun to think maybe of applying straight to a PhD program in Math instead

due to a newly sparked interest in Harmonic analysis, Fourier analysis and related topics (been combing through those free MIT lectures).

Some pros/cons:

I haven't yet taken the Math subject test since I planned on applying to EE. Nearly all require it for admission to Math programs

(surprisingly enough). The soonest available test date isn't until October, and I was hoping to have apps out by this time.

In applying to Math, there would be no 1-2 semester remediation period that I would expect with EE. On the other hand, EE also seems to offer a

more open career path.

Apologies for writing a novel, thanks for any feedback.

Why didn't you just do bachelor's in EE if you need a job?

[mhancock743]

Did I come off as if a degree was a means to a salary? That was not my intention. I began considering EE as an option to research signal processing, as it could potentially combine my two main interests of music and science - the prior I thought would remain as a hobby, and the latter I thought I could consider a career (which is not to say that I don't find math and science equally enjoyable) . Why I'm not in EE and the decisions that led my to my present position are a longer and different story all together. I guess maybe I was unaware of the option? Not everyone has their life pinned down at age 18 going into their undergraduate education.

[maclane]

Not really sure what your question is, but there's absolutely no reason you should worry about the Math GRE not available to be taken until October (although maybe you should worry about preparing for it). You'll get your results back in November with plenty of time to spare. There's really no reason to try to finish your applications by October, as opposed to December/January. There's about no chance your recommenders will have their letters submitted by October anyway, and there's even less of a chance that the programs you apply to will start reading applications that early. You have plenty of time to work with; best not to squander it.

If you're just getting into math recently, you probably don't have tons of classes under your belt, so that could be an issue. It may also make preparing for the Math GRE quite stressful.

[mhancock743]

maclane, yes. That is essentially what I was attempting to ask and have come to find.

With PhD programs being so competitive, I wasn't sure if I would be at a disadvantage waiting until the fall to take the subject GRE.

I'm not new to Math. I'll be graduating this December. The only part I'm worried about is I will not have taken a class in Analysis, which

hopefully with good score on the subject test, could make this less of an issue.

Here's classes I have taken:

Math: Calc I-III, Linear Algebra, Stats, Numerical Analysis, Discrete Structures, Abstract Algebra,

Fundamental Concepts in Math (proofs course), Operations Research

Comp Sci: Python programming(tutored the course four semesters), Two courses in C++ programming, Computer Architecture

This Fall I'll be taking Differential Eq's, Mathematical Modeling, Computer Simulation, and Parallel Programming.

I completed all of the above in 4 semesters because I switched to Math at the end of my sophomore year. Suffice it to

say I was unhappy in a different program for a while, and I have never had second thoughts about the switch.

I have no doubts that I could handle it at the graduate level, but I just have concerns of whether I would be considered as serious

an applicant as others applying.

Damn it looks I went and asked an open ended question again. Oh well

[maclane]

When you take does not matter. If you're nervous about it, then take it in October. This will give you the opportunity for a retake in November. Only a few applications are due before the December hear-back date for the November test, so no worries there. But ETS is annoying in that the sign-up date for November is a few days before the October test, so if you need it just in-case, you have to spend tons of money (and to get your October score before the November test, you have to spend an extra $12 or something getting the score early by phone).

Anyway, the CS will be pretty much irrelevant both for the test and your applications' success. There may be one easy GRE question asking you to follow some algorithm they give you the pseudocode for. For applications, your CS stuff will just basically be an extra line on your resume unless it has significantly impacted your mathematical interests as indicated on your personal statement or your recs, etc.

As for the content of the GRE, there will be a few questions on topology and complex analysis, but trying to rush through that stuff just for the test just isn't worth it IMO. The biggest worry I'd have is your missing real analysis. There are several questions that will be very difficult if you only know the computational stuff (calculus). There's just no way around not knowing analysis well. You'll need it for the GRE, for grad school courses, and you'll need it for the analysis qual obviously. Nevertheless, about half of the test is just speed Calculus, so the number one priority is being very comfortable with Calculus. There's no substitute for working lots of problems (TAing Calculus for two years helped me a lot). There will also be a couple questions on DEs, so focus on that in the fall.

Some schools might expect you to have more courses under your belt, but if you can manage a competitive score on the GRE, then they might not hold that against you.

Edited June 15, 2011 by maclane

[mhancock743]

Yes I already signed up for the October test. I'll probably just sign up for the November one as well just to play it safe and to flaunt my wealth (kidding about that part). I knew the Comp Sci courses wouldn't help for the subject test, but I figured it would help to demonstrate logic and problem solving abilities on the app (hopefully anyway). I'm assuming it will moreso since I'll be applying to applied programs. Anyway, that's a different topic.

It looks like I'll be trying to do a lot of self-teaching as far as Real Analysis is concerned. Know of any good sites or textbooks?

[maclane]

Well I'm not really much of an analysis person, but Rudin is pretty much the standard (I think it's called Principles of Mathematical Analysis). It's not super easy to just read independently though... I think it tends to overcomplicate things by doing stuff like focusing on metric spaces and topology a lot of the time instead of just using the real numbers, which just makes it more of a pain in the ass notationally. I've also used Bartle and Sherbert's book. It's alright, but I don't really like its treatment of integration and limsup/liminf.

[mhancock743]

Yes, I think a multitude of sources is the best way about it. I picked up this one. We'll see if it was worth the $5

I'll definitely look into the ones you suggested when I feel like spending a little more money, and I thank you for your help.