PDA

View Full Version : Imperfect information



the_asker
08-25-2007, 10:09 AM
I suppose it's clear to those who frequent this forum that upper-level math classes are necessary qualifications for admission to top programs. That is, what UPenn's website (Admissions Information to the Ph.D. Program in Economics (http://www.econ.upenn.edu/Graduate/Admissions.htm)) calls an "excellent level" of preparation is, for most of us, effectively the minimal level:

A minimal level of preparation consists of two years of mathematics courses in college, including courses in:
multivariable calculus
probability theory and statistics
linear algebra.

An excellent level of preparation consists of additional courses in:
real analysis
point-set topology
measure-theoretic probability theory

My question is, is this clear to the majority of applicants as well? Or will most of them give themselves the benefit of the doubt and assume that the linear algebra and calculus prerequisites suggested by most school websites might get them through?

As an additional question, which is actually more helpful in grad school, measure-theoretic probability theory or stochastic calculus?

ForTheWin!_08
08-25-2007, 12:08 PM
Stochastic calculus is crucial in financial economics, not so much elsewhere. However, understanding measure-theoretic probability theory is a prerequisite for stochastic calculus.

Furthermore, a measure-theoretic approach to probability theory is basic to research in econometric theory, and micro and macro theory too.

snappythecrab
08-25-2007, 04:29 PM
I suppose it's clear to those who frequent this forum that upper-level math classes are necessary qualifications for admission to top programs.


Neither necessary nor sufficient.

Karina 07
08-25-2007, 04:49 PM
Neither necessary nor sufficient.

Agreed. Helpful, but.

the_asker
08-26-2007, 03:20 AM
So what come closest to sufficient qualifications? And are there necessary qualifications not outlined on program websites?

freelix
08-26-2007, 05:39 PM
Sufficient? I would argue that there's no such thing as having a "sufficient" background for guaranteed admission. All you have to do is scroll down the profiles and results thread from this year to see that many, many stellar applicants don't make it to top schools where they certainly have the ability to succeed. The fact is that there are limited slots at every econ program, and excellent candidates face the possibility of being turned down from any one of them.

I definitely agree with the previous two replies, though, that you're wrong if you think you think having courses in topology and measure-theoretic probability constitute minimal requirements for admission. I have a friend who's starting U Penn's program this year who has taken neither of these--so I guess he doesn't have "excellent" preparation as defined by Penn's website. For another top 10 perspective, here's NYU's FAQ Admissions page:

"It is also important to realize that we are not looking for mathematical wizards but individuals who have a reasonable amount of mathematical maturity and can apply their logical mind to economic questions."

The take-away? Take enough math and do well enough to show you can handle rigor--and here a course like real analysis is a good signal--but don't think you need to have every course in the math and stats department on your transcript to have a realistic chance of getting into a top school.

Necessary? I would say strong letters of recommendation from economics professors who know you and your work well, WANT to see you get admitted to a top program, and can convincingly communicate to the adcoms WHY they believe you deserve to get in (i.e. can vouch for your future promise as a RESEARCHER, not merely as a student who can get good grades in hard math courses). I really think this aspect of the application is absolutely crucial, but is so often overlooked by posters in this forum in favor of the math courses. Again, from NYU's website:

"It isn't easy to define this, but what we look for is evidence of creativity, originality, drive ... ingredients that will make you a successful researcher. The ability to do well in courses or exams is correlated with this talent, and we certainly look for that and we also demand that you do well in our coursework, but that isn't the ultimate goal. Many students are really good at coursework but are just not the same when it comes to research."

I hope I've provided some counter-balancing food for thought on this issue.

SquareSquare
08-26-2007, 06:28 PM
support freelix . no silver bullet (maybe if your & your gf's fathers are econ nobels), no magic formula.

Intuitively, adcoms will try to have a relatively diverse pool, so they won't admit 28 topologists... Maybe doing something different will increase one's chances.

I recall a while ago there was a student that performed as a clown before doing Masters and then going to Chicago. Hmm, that doesn't sound like a good example... (I doubt that had anything to do with his admissions)

Yet, my point is still that a bit of diversity in your courses could be beneficial .

SquareSquare
08-26-2007, 06:36 PM
A small green humanoid in the library.

michaelmas
08-26-2007, 06:37 PM
Many students are really good at coursework but are just not the same when it comes to research."


I hope this is true the other way around for me.

studentecon
08-26-2007, 08:07 PM
I suppose it's clear to those who frequent this forum that upper-level math classes are necessary qualifications for admission to top programs. That is, what UPenn's website (Admissions Information to the Ph.D. Program in Economics (http://www.econ.upenn.edu/Graduate/Admissions.htm)) calls an "excellent level" of preparation is, for most of us, effectively the minimal level:

A minimal level of preparation consists of two years of mathematics courses in college, including courses in:
multivariable calculus
probability theory and statistics
linear algebra.

An excellent level of preparation consists of additional courses in:
real analysis
point-set topology
measure-theoretic probability theory

My question is, is this clear to the majority of applicants as well? Or will most of them give themselves the benefit of the doubt and assume that the linear algebra and calculus prerequisites suggested by most school websites might get them through?

As an additional question, which is actually more helpful in grad school, measure-theoretic probability theory or stochastic calculus?



As snappy (well) said, these are neither sufficient nor necessary. I guess one sufficient condition is to have a paper in Econometrica, AER, JEP, QJE or RES :D.

A weaker sufficient condition is to be the best student in the undergrad economics program of a well known university.

As for the maths, clearly the more the better, but since we are economists the real question is 'what's the opportunity cost of taking one extra math course and what's its marginal benefit?'