3.47 GPA am I screwed for Stanford Math PhD?

[donnylee]

Dear Physics Forums,

I am a math major who just finished my sophomore year at a USNews top 10 school. As the title says it, I have a (sad) 3.47 gpa with my math gpa just about the same. Over the last few weeks, I have been considering my career path and yes, going to graduate school for math is my 80% favored option, and the school I'm aiming at is Stanford possibly doing research in differential geometry or stochastic calculus.

Yeah yeah, I'm one of those get a math PhD and then transit to Wall Street kindda of person. But that's not the subject of this post and definitely not one about why I'm studying geometry if I want to go into finance. I have other reasons for a math PhD and just take it for now that I want a Math PhD tag on my resume, no choice about that. Specifically, this is an inquiry of my chances of getting in Stanford, preferably applying after my fall semester junior year. Yes, it's fast and I'll skip the details of the rush. So my profile, and by no means showing off (lol, there's nothing to brag about really), to give you a better picture of where I stand.

Relevant Math Classes
Undergrad level: Advanced Calc (B), Elementary ODE (A), Math Seminar (A-), Probability (B+), Numerical Analysis (A-)
Graduate level (as I know this is more important)
(B to B+ grade) Algebraic Forms, Topology, Mathematical Finance
(A- to A grade) Stochastic Calculus, Mathematical Modeling

Relevant non-Math Classes (to show ambition and intellectual curiosity)
Quantum Information Science (B+)

Others
Research Position at Nanophotonics Labs doing Matlab programming. Certainly not high-caliber math but it could lead to a publication.
2010 Honorable Mention in the MCM.


So here is my plan. Some of the non-A grades can be attributed to a busy semester, that is one which I participated in various school musicals, students groups, entertaining girls. But now, next fall is all business and I'm adopting an attitude of win or go home. Hence, the following is my projected plan and those pesky assumptions which of course have large variances. Don't worry about enrollment requirements in classes because my school is liberal in this regard.

1. Enroll in 4 math grad level classes.
Functional Analysis (to redeem my B in Advance Calc), Representation Theory (another redemption class), and Elliptic PDE or Algebraic Topology. And one independent study in Stochastic Calculus. Assume for now that I get A+s for all. (Still a 4.0 in my school though)
2. Enroll in 2 non-math grade level classes. These are my interest and MAY show diversity in knowledge. Assume again A+s.
CMOS VLSI Design and Advance Computer Architecture.
3. Take part in Virginia Math Competition. Assume say top 20.
4. Try to submit an accepted publication preferably in an area I'm most familiar with. Possibly a product of the independent study.
5. Release an iPhone app which is a numerical solver for PDEs. (my side project which I'm working on now)
6. Perfect GRE, at least for math. Failure is not an option.

Things not going my way are.
1. I'll probably have a 3.58 gpa upon application submission, definitely not in the league of those 3.9s.
2. I have no past results in any mathematics individual problem solving competitions, that is the Putnam.
3. I'm not sure whether the admissions committee will buy taking higher level classes to cover the weaker lower level ones.

So let's just say that all I accomplish all that I set out to do in the fall ... and ... I take a stab in the dark for Stanford (and NYU), I need to know where I stand? I understand that no certainty can be guaranteed. At least if I know how this forum feels about me, I have a target to aim towards in the fall as opposed to working towards something which is nothing but a loss cause.

I appreciate any feedback in this matter.
Thanks.
Lee

 

[kramer733]

I wish i got grades like you.. =/

 

[Academic]

Surely there are other schools besides Stanford that do research in the area you want. Frankly, I don't see the point in shooting for a specific school (unless you have some other family reason, etc). Shouldn't you be more interested in specific research advisers?

 

[donnylee]

I am aware of the other schools which for the sake of argument, let's explicitly label them as Stanford, Caltech, UChicago, NYU, Columbia and UPenn. I know that in search for a graduate school, the specific advisor may carry more weight than the school, however I seek brand name and advisor.

I was just using Stanford and a microcosm for these schools and the purpose of my enquiry as to see whether a 3.58 gpa person and assuming with all of the above, stands a chance getting through the door of those schools.

Thanks.

 

[donnylee]

Oh, also add Berkeley to the list. :)
I love the campus and the history of the school too.

 

[Monocles]

You would probably need some very, very strong letters of recommendation to get into those with a GPA like that. Most of these schools you're looking at could fill up their entire class with people who got straight A's in their math courses. A small handful of these 'top' schools (I don't recall which ones) do list the average GPAs, GREs, etc. of admitted students, and I recall the average math GPA tended to be in the 3.9 range. One random thing that has me puzzled is that you mention applying after fall of your junior year. Are you aware that very few schools (certainly none of the ones you listed) accept applications during the spring? Most people intending to start grad school the fall after their graduation apply during the fall of their graduating year.

Obligatory disclaimer: I'm no admissions officer, and actually I am an undergraduate, but I have spent what is probably an unhealthy amount of time reading up on this stuff since I'll be applying to math grad school in a year or so too.

 

[Academic]

Yea, I too would suggest a more well rounded grad. school application list.

 

[donnylee]

Thanks for your reply. I am indeed applying to be admitted in a program which begins in the fall after my junior year. I mean to indicate the fact that I have to submit my application by the end of fall; the deadline is usually end of the year to be admitted in the program next fall, hence I only have one remaining fall semester to prove my case or go home.

The 3.9 average GPA does scare me. With my ~3.5 math gpa right now, I can only hope to have it to 3.65 by the end of this coming fall and that's the number the admission department will see.

I guess the "redeem yourself by A's in more advance classes" don't work?

 

[snipez90]

First of all, I'm curious about the courses you've listed. Graduate topology, where most of the students are presumably first year grad students, should be a really difficult course at any graduate math department in a USNews top 10 school (which I'm guessing includes the usual top schools that many people could list off the top of their heads). If you took that without taking an undergrad course in topology, then that just seems stupid. Of course topology is considered by some to be a graduate level topic, but many colleges offer topology as an undergrad course, probably at the level of Munkres or something similar. Thus I am simply curious as to whether your first topology course was taken with mostly graduate students at the graduate level. Also what the heck are algebraic forms. I'm assuming you mean something other than homogeneous polynomials, which is what wikipedia labels them as.

In any case, the "redeem yourself by A's in more advance classes" probably doesn't work for multiple reasons. It seems reasonable that graduate schools want to see how we'll you're prepared to handle their graduate level courses. Receiving B's in a couple graduate and undergrad courses at your own institution doesn't help, even if you could get A's in your upcoming classes.

Also, I think it's fair to say that obtaining 4 A's in 4 graduate courses is not an assumption to be taken lightly. Even in a "do or die" situation, there are many things that could go wrong. No one knows what you covered in your advanced calculus, but if it's not at the level of say Rudin's Principles of Mathematical Analysis, functional analysis is going to be very difficult. In many cases, a more difficult real analysis covering Lebesgue integration and Lp spaces is usually a prerequisite to functional analysis. Similarly, algebraic topology is likely going to be much more difficult than your previous topology course. The basic issue is that, unless you failed to mention all your math classes, the prerequisites don't seem to match some of the classes you're planning on taking.

As a final note, receiving a B+ in probability and an A of some sort in your stochastic calculus course is going to raise some flags if you plan on doing work in stochastic calculus. Also what does studying stochastic calculus independently entail? Haven't you taken a graduate level course in it already?

 

[donnylee]

Hello snipez90,

Thank you for your kind reply. Some of your opinions did add value to my consideration as to what I will be doing next. It does seem that my profile and path of admission is rather random base on the courses I have listed. You did get half of that correct. Let me quickly clarify.

If you took that without taking an undergrad course in topology, then that just seems stupid. Of course topology is considered by some to be a graduate level topic, but many colleges offer topology as an undergrad course, probably at the level of Munkres or something similar. Thus I am simply curious as to whether your first topology course was taken with mostly graduate students at the graduate level.

Yes, I took it with a bunch of undergrad students (half point-set half algebraic and we used Munkres) but due to some misnomer, it was listed as a graduate level course. Perhaps but not entirely a stupid decision to take it as 1. I did have some advance calc background and 2. it exposed me to what topology is.

In any case, the "redeem yourself by A's in more advance classes" probably doesn't work for multiple reasons. It seems reasonable that graduate schools want to see how we'll you're prepared to handle their graduate level courses.

If what you say is true, then I guess there isn't a second chance for me to show my aptitude in math?! That's really a bummer. I know that undergraduate mathematics is a way to filter out students who are not capable to do graduate level math, but do I get a second chance? And to me, taking higher grad level classes (though seems stupid and haphazard) is the only way.

Also what the heck are algebraic forms.

Sorry I meant algebraic structures, a fancy name my school uses to label graduate level abstract algebra. We did use Dummit and Foote first third of the book and some chapters on fields.

The basic issue is that, unless you failed to mention all your math classes, the prerequisites don't seem to match some of the classes you're planning on taking.

Hmmm, are you telling me that graduate school admissions also look at the order of classes you take in addition to what 'end-track' classes you take? I would believe this to be the case, however there is some allowance for me and other math students to have taken classes over the summer at some random place which gave them training in a certain area. This would not be listed in the transcript but sometimes this training does serve enough, again, to take the 'end-track' classes.

From your reply, I sense a tone describing the futility of my plan. Indeed, you could be correct and any admissions officer could feel the same way. Nonetheless, I need to think of something even if it means a do-or-die endeavor in my fall semester. The good news is that I do have some time on my hands, 2 months before fall term, to prepare for the semester leading to the application deadline.

P.S.: And if it all fails midway through the fall, I could withdraw from half my classes, take a bunch of easy classes in the spring, get a 3.6 (reasonable) and work at some C++ programming shop. :-)

 

[donnylee]

Oh and sorry to mention that my Probability and Stochastic Calculus was very different. It was Stochastic Calculus from a analysis point of view, about (65% measure theory and 35% application), so a B and A respectively is mildly reasonable.

And certainly one graduate level course doesn't cover all of Stochastic Calc. An independent study here usually means 1. a more indepth study aka Malliavin calculus or 2. research topic.

Any feedback from you is greatly appreciated.

 

[twofish-quant]

There is an extremely high degree of randomness with graduate school admissions, and a 3.47 GPA makes admission to Standard math unlikely, but not impossible. On the other hand, you could have a 4.0 GPA, and still not getting admitted for twenty different reasons.

I really think that you are going about your academic planning badly. Because there is a high degree of randomness and luck in the system, at some point the dice will go against you. You really should not be counting on getting into Stanford or any other particular school, but rather do your planning so that you can get what you want assuming things don't go they way you won't, because at some point, they will go in another direction.

Let me ask you a more relevant question. Suppose you found out that it *was* impossible for you to get into Stanford, if not for GPA, but for some other reason. What do you do then?

 

[twofish-quant]

Yeah yeah, I'm one of those get a math PhD and then transit to Wall Street kindda of person.

If that's what you are really interested in doing, then you are probably better off getting your Ph.D. in applied math or statistics. One thing that you have to be aware of is that the status hierarchy on Wall Street is very different than it is in academia. Basically, in academia you get more points the more abstract and esoteric your research is, whereas on Wall Street, you are more points the *less* abstract and esoteric your research is.

The other thing is that mathematicians and physicists on Wall Street typically *don't* came from big name schools. People from big name schools are more likely to go into academia, so the people that do end up on Wall Street tend to be from schools that are decent but not ultra big names.

I have other reasons for a math PhD and just take it for now that I want a Math PhD tag on my resume, no choice about that. Specifically, this is an inquiry of my chances of getting in Stanford, preferably applying after my fall semester junior year.

Personally, I think you are setting yourself up for a big mess. As the saying goes, you can't always get what you want, but if you try real hard, you might get what you need.

 

[snipez90]

Ah okay, well it's not stupid then. It's just weird to see that someone would list a graduate topology course with an advanced calculus as the only remotely reasonable prerequisite.

I'm an undergraduate, so I'm strictly speaking from a reasoned point of view. I was simply stating that if you handed that in as a transcript, it certainly looks a bit odd. Taking courses at a random place sounds harmful. How do you know that the course you took is as suitable as a prerequisite the one your institution offers you? Also I don't understand why the "middle-tracK" courses don't appear on your transcript. Didn't you receive a grade for these courses? After all they do have to test whether you're ready for the "end-track" courses don't they?

The primary reason I thought your transcript was odd is the lack of analysis. Most people take a yearlong analysis sequence, so it's going to look weird if you take advanced calculus (which in a lot of cases is basic analysis, or an easier version of the "real analysis" course), and then the subsequent courses in difficulty were graduate level courses. I would expect at least one or two analysis courses to fill in the gap between advanced calculus and then a measure-theoretic stochastic calc course. Also was complex analysis another course you took on the side?

As for your plan, I'm not saying you don't get a second chance. At my school, our first year graduate program is very difficult, and undergraduates have to receive permission to take graduate level courses. This typically occurs after they have taken yearlong honors sequences in algebra and analysis, then quarter courses on complex, topology, differentiable manifolds, algebraic topology. There are students who have taken the courses mentioned, received A's in all of them, and were still denied permission because a single graduate course could damage their transcripts. Graduate courses are serious business and I would consider this carefully instead of thinking of it as a "do or die" situation.

I know you've developed a "nothing to lose" attitude, but this is not the best way to redeem yourself. It's one thing to try and increase your GPA, but it's something different when you take things to the extreme and try to do 1-6 in the fall. It is not a reasonable list of assumptions, and the most prepared math major could fall short of accomplishing 1-6 (and you're up against A LOT). So to actually answer your original question, I don't think you're in good standing for Stanford. I do think you could certainly make a good effort towards developing a more realistic list of goals and set out to achieve them. But I can tell you right now that you are not going to get the hopeful reply you might have wished for from other members on the forum, and if this persuades you to carry out your plan outlined in the P.S. now, then I think I have helped save you 4 extra years of trouble.

 

[donnylee]

Hello twofish-quant. Thanks for your contribution to my dilemma. My answer to your more relevant question "who would I do after college?" is simple; and that is to go into the field which you yourself is in.

The most desired (but NOT only) career path for me is Quantitative finance / developer, more specifically using quantitative and computational methods to take advantage of inefficiencies in the market and then trade to make a profit. After reading extensively the forums on quantnet, the minimum qualifications seems to be MSc or PhD and since MSc is terminal and expensive, the decision was simple, get a PhD. I have thus eliminated MSc from the equation.

So PhD at where? Well, another simple answer, top schools. Yes, I do know that my profile as-it-is doesn't cut it. Which is why I'm conjuring an all-in move in the fall.

Now as to your argument of more applied more points in Wall Street, I fully agree. But again, after extensive reading on the forums, the consensus I'm getting is that those coming from 1. a big name school, 2. a background of largely computational work, 3. a thesis which proves you can speak stochastic calculus (analytical or applied) stand the best chance.

This information therefore defined my all-in move: at least in Stanford I get 1 and 3, and NYU, I get all 3. Needless to say, getting 1 first is the top priority.

The other thing is that mathematicians and physicists on Wall Street typically *don't* came from big name schools.

I acknowledge that you are someone in the field. However, my disagreement with this statement comes from the fact that most job descriptions I read regarding this field stress 'PhD in top 10 school, Ivy-school, etc etc.' Unless this is just a convention to attract only top talent, I urgently seek to enter a PhD program in a top 10 school so as to increase my chances on this quant job.

I know I do seem to come across as a stubborn fool who doesn't know the trouble he is getting into. Pardon me then. Nevertheless assume I have the following to work with:
1. One semester to do anything to get into a top PhD program.
2. No $70k to pay for a Masters program.
3. Have a spring semester to graduate with 3.65 and a degree in Computer Science and secure a related job after graduation.

Do I still seem that unreasonable with my endeavor? It's only recently that the flame to join the league of market modelers, quantitative developers and quantitative analyst was kindled and seeing that time is of the essence, I'll be sacrificing a lot in the fall working towards something which I hope is not a loss cause.

 

[fasterthanjoao]

Do I still seem that unreasonable with my endeavor?

Yes.

The points twofish made in the previous post about aiming as specifically as Stanford (or NYU) remain valid - it's a bad way to go about it, will probably set you up for disappointment and could end up meaning you lose a year to the system.

Aiming for a specific school because 'after extensive reading..i've seen in job applications it says..' is just an awful idea. A PhD is an extremely serious piece of a work - you will have some very difficult and trying times during the course of a PhD - you need to choose the school that fits best for you. Part of this is about what you'll get in terms of career needs, of course, but a large part of it is about who you are - which you seem to have totally neglected. You're focussing on a single part of a future job application based on a PhD you not only haven't started but might never get.

In any case, obviously it is always good to aim high - but going to a so-called 'lesser' school isn't going to necessarily harm your chances of finding a job afterwards so don't let that single factor consume you.

 

[twofish-quant]

Hello twofish-quant. Thanks for your contribution to my dilemma. My answer to your more relevant question "who would I do after college?" is simple; and that is to go into the field which you yourself is in.

That's curious, since finance happens to be something that I stumbled into, and it's not my first (or even my fifth choice) of career. It seems interesting, but I think I'll probably get bored of it in a few years and do something else.

Something to point out is that my current job didn't exist when I graduated from college, and I doubt it will still exist in ten years. If you are looking for careers in jobs that currently exist, you are really limiting yourself.

The most desired (but NOT only) career path for me is Quantitative finance / developer, more specifically using quantitative and computational methods to take advantage of inefficiencies in the market and then trade to make a profit.

That's not what most quantitative finance people do now, and I really doubt it will be what people will be doing in a decade.

After reading extensively the forums on quantnet, the minimum qualifications seems to be MSc or PhD and since MSc is terminal and expensive, the decision was simple, get a PhD. I have thus eliminated MSc from the equation.

I don't think that's a good decision process. I really think that you need to think some more about what you really want. Money? Power? Social approval?

So PhD at where? Well, another simple answer, top schools. Yes, I do know that my profile as-it-is doesn't cut it. Which is why I'm conjuring an all-in move in the fall.

The trouble is that you may not get into a "top school". Then what? For me, it's pretty simple. Since I think physics is cool, if the only place I can study physics is in a garbage dump, then I'll study physics in a garbage dump.

Now as to your argument of more applied more points in Wall Street, I fully agree. But again, after extensive reading on the forums, the consensus I'm getting is that those coming from 1. a big name school, 2. a background of largely computational work, 3. a thesis which proves you can speak stochastic calculus (analytical or applied) stand the best chance.

I think you are very strongly misinformed. It should be pointed out that a lot of the people in the forums have no idea what they are talking about. Also, it should be noted that schools sometimes vastly overstate the importance of a big name so that they can sell their MFE programs.

The amount of stochastic calculus that you need to do most jobs on Wall Street can be written in five pages, and if you have a decent math background, you can pick it up in two weeks.

This information therefore defined my all-in move: at least in Stanford I get 1 and 3, and NYU, I get all 3. Needless to say, getting 1 first is the top priority.

So what if it doesn't work? What if the only school that you get into is North Podunk? What do you do then?

I acknowledge that you are someone in the field. However, my disagreement with this statement comes from the fact that most job descriptions I read regarding this field stress 'PhD in top 10 school, Ivy-school, etc etc.'

Job descriptions are bogus. What happens is that the hiring manager goes to the head hunter who puts out the same standard job ad. Job ads never tell you the real qualifications. They are just there so that HH can get resumes, which they then filter.

Unless this is just a convention to attract only top talent, I urgently seek to enter a PhD program in a top 10 school so as to increase my chances on this quant job.

So why do you want to be a quant anyway? IMHO, quants don't exist.

Do I still seem that unreasonable with my endeavor?

Extremely.

It's only recently that the flame to join the league of market modelers, quantitative developers and quantitative analyst was kindled and seeing that time is of the essence, I'll be sacrificing a lot in the fall working towards something which I hope is not a loss cause.

You are going about it all wrong. Spend some time trying to figure out what you enjoy. Learn some things. Try to figure out *why* you want what you want.

 

[JCVD]

Unfortunately you will not be competitive for admission into any top schools with a math GPA below 3.7 (really, below 3.8). It would also be in your best interest (in terms of making yourself a better applicant) to stay 4 years as an undergrad, but if there are prohibitive circumstances out of your control I guess there's not much you can do. (You might want to mention those circumstances on your applications so that schools know you aren't just trying to take a shortcut. Unless those circumstances involve you expecting to get kicked out of school when your moneymaking scheme of stealing prescription drugs from senior citizens and selling them to terrorists gets discovered. Then I wouldn't put that on my app.). Also, you should try to take a course in complex analysis before you graduate; grad schools look for that course specifically.
If you are wondering what types of applications will get accepted where, this might help.
http://www.mathematicsgre.com/viewtopic.php?f=1&t=357
Finally, know that getting a PhD from a top ten school is important if you ever hope to hold a decent position in research academia, but getting a PhD from a lesser school will not hold you back so much from nonacademic jobs. Since you definitely want to go to grad school, I think it would behoove you to not throw all your eggs into the "I'm going to be able to grossly outperform my previous academic record this fall even though I'm taking 4 grad courses whose prerequisites I didn't do so hot in all while studying for the math subject GRE and trying to rush a paper for publication (which, even if you are somehow able to submit in the fall, will in all likelihood not be responded to until well after schools are done making decisions on grad applications)" basket, since if things don't go well, you could be looking at math GPA even lower than the 3.5-3.6 range that you aren't even in yet but that still remains very borderline for even garnering serious consideration for admission to any school in the top 70, let alone top 10.

 

[donnylee]

Good evening to all the kind people who have replied to my post.

I thank you for your feedback and appreciate your thoughtfulness in advising me what to do. I have read, and reread, everyone sentence, analyzed it, and have reached this final state.

I stand corrected. My all-in fall semester is ludicrous. I'll certainly revise my fall schedule. To what? I do not know as of now. PhD or no PhD, quant or no quant, I need to cut back my losses right now and seek out a plan which is more doable and secured in terms of my abilities and future destination.

Thank you once again.

 

[twofish-quant]

Finally, know that getting a PhD from a top ten school is important if you ever hope to hold a decent position in research academia, but getting a PhD from a lesser school will not hold you back so much from nonacademic jobs.

In astrophysics, it's useful but not vital to get a Ph.D. from a big name school. The name of the thesis advisor is more important. I've been told that things are very different in math.

Something that you will have to realize is that things that will make you better at dealing with industry will make it harder for you to get an academic position. For example, in industry, you are better off having a broad view of things, whereas in academia you end up being a narrow specialist. Something that seriously hurt me in graduate school admissions was the fact that I learned a lot of computer programming and Chinese history. Those have been massively useful in industry, however.

The other thing about Wall Street quants, is that physics and math Ph.D.'s generally *don't* generally get their graduate degrees from big name schools (with MIT being an exception), because finance is just not the first choice of careers for physics/math Ph.D.'s. If you get your Ph.D. from Harvard, you are going to likely be able to get a research position somewhere. If you get your Ph.D. from North Podunk, you aren't, and making tons of money on Wall Street is a consolation prize. You see a lot of Harvard MBA's running around Wall Street, but I've never met a physics/math Ph.D. on Wall Street with a Harvard or Stanford Ph.D.

Personally, I think this makes the environment more friendly and tolerant than academia or even finance Ph.D.'s or MBA-dominated fields. One reason that Harvard physics Ph.D.'s don't end up on Wall Street is that having a physics Ph.D. from Harvard doesn't give you anything over one from North Podunk. (MBA's are *VERY* different.)

Something to remember is that the US graduates about 1000 physics Ph.D's each year. Harvard alone produces 1000 MBA's a year, and the total number of MBA's that are graduated each year is 100,000.

You could be looking at math GPA even lower than the 3.5-3.6 range that you aren't even in yet but that still remains very borderline for even garnering serious consideration for admission to any school in the top 70, let alone top 10.

The other thing I'd seriously suggest is to spend some time thinking about *WHY* you want what you want.

One of the basic human motivations is getting attention and approval from other people, and something that top students get in high school is a lot of attention and approval. You pass the test, get good grades, and then you get all these awards and people talk about you. So the game in high school, is to get A's on everything and get the awards.

Once you get into college, the game is different. One problem is what I call the "video game level" problem. What happens if you are the top 5% is that you move to the next level. Maybe you will be the top 5% of that, but if you are, then you move on to the next level. The thing about this system is that eventually you will end up either average or (gasp) below average! After being at the top 5%, it's a shock to find yourself in the middle 50% or worse yet at the bottom 5%. Dealing with this fact is something that pretty much everyone has to deal with eventually, and dealing with it early is better than dealing with it later.

The other big difference in college is that you have choices and no one tells you want the game is. High school is pretty structured. They hand you the test, you do well, you get the prizes. The thing about this system is that you don't really have to *think* about what is going on. You just take the test. It's something of a shock when things are less structured and defined, it's not clear what the rules are, or there are multiple conflicting games.

If you want to get your Ph.D., that's great. But you have to figure out for yourself why you want to do it. If you want to make it big on Wall Street, getting a physics/math Ph.D. is probably one of the absolute worst ways of doing it. You can get equal amounts of money and success with *much* less pain and agony. So getting a physics/math Ph.D. for the purpose of getting a job on Wall Street is a truly bad and stupid idea.

 

[twofish-quant]

One big mistake that math/physics majors that want to go into finance make is to try to focus on topics that they think are "relevant" to finance. This is a seriously bad thing to do because the information that people get as to what is "relevant" tends to be old and outdated.