little math, but As in PhD courses

[sulebrahim]

There are various schools where MA/MS students can take the PhD sequence.

 

 

 

In some of these schools, the MA programs admit students with only calc1. The students are advised to do some self studying by themselves, and some have actually succeeded by getting As in the micro, macro and metrics sequence and other field courses as well.

 

 

 

How does the lack of math on your profile get looked at?

 

 

 

I'm not asking about my case, just curious, because I know a few students who do well and have PhD aspirations.

[Food4Thought]

Most of the PhD students at my school (top 40) have one of the two backgrounds: 1) a master's degree in economics but not much formal math, or 2) no master's degree but at least a math minor's worth of coursework.

 

I think that, at most mid-range schools, this is the case. At higher ranked programs, however, I can't imagine that getting good grades in your master's degree would be able to replace mathematical preparation. That being said, this is pure speculation. I bet there are more informed opinions on the matter. :playful:

[Catrina]

I don't even see how it would be possible to do well in the PhD core classes with only having taken calc 1, unless the person somehow taught him or herself a substantial amount of math.

 

Food4thought, while it certainly seems that one can get away with less math at mid-ranked schools if there are enough other positive factors, I think that you may be using "not much formal math" differently than Sulebrahim is. Are there really students at your school who haven't taken calc 3 or linear algebra?

 

Even among similarly-ranked schools, the degree to which math background matters (for success, and probably for admissions too) seems to vary by program. At one school that I visited, the students said that the problem sets (and exams IIRC) were primarily proofs, while at another very similarly ranked school I was told that there were very few proofs on problem sets and exams.

 

Also, keep in mind that admission isn't the only thing that matters. Real analysis is actually used in first year coursework, and you don't want to be trying desperately to learn RA during a short math camp review or while trying to do the actual coursework

[sulebrahim]

I think you have a point Catrina. I think at some schools, there are rarely proofs in the problem sets or exams and the tests are more applied. So getting an A in Micro 1 and 2, does not entail one would succeed elsewhere.

 

I just thought it was interesting how some people in my undergrad and current school were able to get in as masters students, catch up, and then stay on for PhD as opposed to having to take math and then apply somewhere else.

[fakeo]

I don't even see how it would be possible to do well in the PhD core classes with only having taken calc 1, unless the person somehow taught him or herself a substantial amount of math.

 

I think you answered your concern yourself. Self-teaching math is entirely possible. Let's not delude ourselves, the math requirements in econ are not easy, but are doable with a year or so of targeted self-study (with Simon and Blume or similar) if one has some exposure to calculus (like from high school) and perhaps some exposure to linear algebra. After that, acing PhD-level courses is entirely possible.

 

Anyways, I think sulebrahim raises an important point, especially because it applies to me. :) I did little formal math in undergrad, and after some rigorous self-study of about a year, I could more or less ace PhD-level core courses. So I'm very interested in how adcoms are gonna view this come admission season. I think doing well in PhD-level core courses should be sufficient proof that a person has enough math skills for an econ PhD, after all we wanna be economists not mathematicians. But obviously I have a conflict of interest here.

[Food4Thought]

Anyways, I think sulebrahim raises an important point, especially because it applies to me. :) I did little formal math in undergrad, and after some rigorous self-study of about a year, I could more or less ace PhD-level core courses. So I'm very interested in how adcoms are gonna view this come admission season. I think doing well in PhD-level core courses should be sufficient proof that a person has enough math skills for an econ PhD, after all we wanna be economists not mathematicians. But obviously I have a conflict of interest here.

 

I think it's a healthy rationalization to make, definitely better than doubting yourself and your abilities because you didn't take more math. However, I'd be interested in hearing other's opinions on how well this tactic would work for upward mobility. I am skeptical.

[to2012]

never mind.

[Insti]

Just to second Catrina's point - math requirements vary a lot by program. I feel you should be fine with acing PhD courses and no math given that the program where you aced those classes has a solid reputation of having at least as rigorous PhD sequence as the ones you are applying at. It is true that there are schools where you can pretty much get away without writing proofs all night long and that could be a minus when you apply, but remember that a PhD class in micro or macro or metrics is more than a math class. Acing a PhD class would signal good mathematical ability and a solid understanding of the economics that goes with it, something that a math class cannot give you.

[fakeo]

It is true that there are schools where you can pretty much get away without writing proofs all night long and that could be a minus when you apply

 

Yes indeed, it may depend on the rank of the school where one did the PhD courses. If you can excel at PhD courses at a particular school, then this should at the very least show adcoms at similarly ranked (or weaker) schools that you can handle their PhD coursework. Of course, this is just how I think things should be, not necessarily how they are.

[to2012]

I really have to think that unless you have some truly-special letters of recommendation or the doctoral courses that you've "aced" were taken at a highly-ranked institution, not having Calculus 3 as an absolute bare-minimum is going to hurt an applicant substantially. We're mostly all familiar with Levitt's story but that's clearly an atypical one. Calc III and linear algebra are such easy classes that anyone who is able to achieve the postulated level of performance in a graduate microeconomics course should be able to complete them with very little effort.

 

It seems like applying without satisfying the lowest level requirements is a roll of the dice.

[fakeo]

Still, I don't see why the fact one can handle PhD-level courses at a similarly ranked institution is not enough proof of the applicant's mathematical abilities.

 

Probably it's different in the US, but taking Calc III or other random courses as a non-degree student or even as an elective in an econ masters is borderline impossible in the educational system I come from. So this was never an option for me.

 

Anyways, yes to a certain extent it's gonna be a roll of the dice. But I'll take a shot at it.

[Food4Thought]

The director of graduate studies at my school put it this way:

 

"Everybody has a limit. If a student at a lower ranked school takes all the required classes and gets all A's, it doesn't really tell me if they can handle much more or if they have met there limit and just skated by."

It is for this reason that taking a PhD course to prove your mathematical ability may be less effective than taking a lot of math. You may have just scraped by in the PhD econ course with that A, or you may have thought it was a piece of cake and easily secured your A. Ideally, this is where a letter of recommendation would come into play, but sometimes it doesn't, because recommenders might be wanting to fudge an applicants ability a little bit (a la Crawford-Sobel) so as to get them into a marginally better program. But if you have many A's in lots of math classes, this uncertainty regarding your abilities becomes less severe.

[fakeo]

I would say getting A's in grad courses is a little bit more than just taking "all the required classes". Taking PhD core courses at a similar place to the one you're applying to is almost equivalent to having already completed the first semester/year of that particular program successfully. It is as good/direct a proof as you can get of one's ability to handle advanced econ coursework.

 

And what if grad micro is your limit? If one is able to handle core courses, then beyond that what should matter is research ability, no?

 

Anyways, I was just stating an opinion. I agree that if an applicant wants to maximize admissions chances, then they should take a lot of math, etc. I was just trying to come up with some reasons for why adcoms should have love for some people with alternative backgrounds.

[Econhead]

I would say getting A's in grad courses is a little bit more than just taking "all the required classes". Taking PhD core courses at a similar place to the one you're applying to is almost equivalent to having already completed the first semester/year of that particular program successfully. It is as good/direct a proof as you can get of one's ability to handle advanced econ coursework.

 

I think your question points back to why programs make transfer students (even from a comparable or arguably better institution) retake courses that they have already taken. Whether this is in the rare case of a Ph.D transfer or an undergrad transfer. The concern, I think, is that "it's not THIS institution." No matter how thorough the syllabus, it's always a concern of whether the coursework is the same and thus whether the preparation for future coursework is the same. I think this is a concern based in wanting students to succeed in the program more than anything else.

[Insti]

And what if grad micro is your limit? If one is able to handle core courses, then beyond that what should matter is research ability, no?

 

 

That is correct if a grad micro course as a standardized commodity. The thing people are saying is that acing a grad micro course in one school does not mean you will not fail the same course at another school. Unless there is a very strong belief the grad micro course you took was more math heavy than the other, it will not send a reliable signal for your math skills.

 

 

This is even the case for similar ranked schools. There are certain people in my program who transferred from other schools and they were saying how different courses could be. Even though the syllabus might say you covered topic "X" it may very well be the case you covered it in much less detail. Also, there is a great difference between the exams - questions like "Check property A applies to choice correspondence X" is not the same difficulty as "Prove property A is equivalent to property B in the family of correspondences X"

[fakeo]

That is correct if a grad micro course as a standardized commodity. The thing people are saying is that acing a grad micro course in one school does not mean you will not fail the same course at another school. Unless there is a very strong belief the grad micro course you took was more math heavy than the other, it will not send a reliable signal for your math skills.

 

I see. I had no idea the gap between even similarly ranked schools' grad courses could be so significant.