PDA

View Full Version : graduate micro or RA



rob59404
05-17-2009, 06:22 PM
Hello everyone, I have another class conflict similar to the one I asked about two weeks ago http://www.urch.com/forums/phd-economics/115505-class-question.html (http://www.urch.com/forums/phd-economics/115505-class-question.html). I can either take graduate micro or undergraduate analysis which has this course description from our catalogue:

UG 451 Introduction to Real Analysis 4 cr. Offered autumn odd–numbered years. Prereq., MATH 251, 305. Theory of metric spaces and point set topology, Riemann–Stieltjes integral, sequences and series of functions. Stone–Weierstrass theorem, theorem of Arzela–Ascoli, introduction to Lebesgue integration.

The micro course is pretty standard. I think the professor uses either Varian's or Nicholson's graduate theory text. I am also taking probability, diff eq, and an applied stats course, so I am a little reluctant to add another math course, but I think I could handle it. Another thing to consider is the intelligence of my class mates. The RA class has only four people registered for it, all of whom are math majors with more experience than me. The econ class will have kids who don't know what the chain rule is, so it will be easier to stand out. Finally, the RA class suggests taking an intro to proofs course, which I haven't taken. I have some limited exposure to proofs from linear algebra and a logic course through the philosophy department. I reason however, that I can self study an introductory abstract math text and pick up the main themes. So that is my story. All advice is appreciated.

economics
05-17-2009, 06:49 PM
GENERAL RULE on the trade-off between math and grad econ: the crappier your university, the less the adcom trusts your econ grades. while econ classes vary a lot by university (quality of instruction, rigor, material covered, competition against classmates, who may be geniuses or idiots), math classes are rigorous anywhere.

therefore, if your UG institution is not top40 in econ, take the math class! it will have higher signaling value.

(that being said, the question is: do you NEED signaling? the answer to the first question answers the second one.)

economics
05-17-2009, 06:52 PM
PS: " The econ class will have kids who don't know what the chain rule is, so it will be easier to stand out." may answer your question - if that's the case, then the econ class is clearly worthless in terms of signaling. that being said, you may become a better economist by taking more econ, obv.

mazzalupi
05-17-2009, 06:53 PM
At most top schools, everyone in the graduate microeconomics class will also have had at least one semester of real analysis as an undergraduate or covered it in your school's summer math preparation class for econ students. So, if you haven't had it, you might be at a disadvantage and it may be hard for you to do well in the graduate microeconomics.

With that said, I think that math is sometimes overemphicized at the expense of taking real economics classes. Taking more econ classes may be better for a number of reasons (including greater contact with economists leading to either research opportunities, letters of recomendation, or just advice on graduate school).

That the graduate micro text is Varian or Nicholson makes me think it may not be a Ph.D. class but another type of graduate class (also your comment about the chain rule). If this is the case, you'll probably be fine without analysis.

Real analysis is important but by no means required for admission to a top school. Most people will have it, but if you are strong in other areas you may be fine.

Zmoney
05-20-2009, 12:29 AM
GENERAL RULE on the trade-off between math and grad econ: the crappier your university, the less the adcom trusts your econ grades. while econ classes vary a lot by university (quality of instruction, rigor, material covered, competition against classmates, who may be geniuses or idiots), math classes are rigorous anywhere.

therefore, if your UG institution is not top40 in econ, take the math class! it will have higher signaling value.

(that being said, the question is: do you NEED signaling? the answer to the first question answers the second one.)

The one thing i do have to add on this is it depends on your realistic target. If you feel like you are going to make it into a top 20 program then Economics advice probably applies the best. Below that however i think the value of RA is just less in comparison to grad micro regardless to program strength.

The OP in a previous thread said they went to a mid ranked state school so probably not top 40 in econ

rob59404
05-20-2009, 03:10 PM
Thanks for the responses. After taking what everyone said into consideration I've developed the following game plan:

1) I'll register for grad micro in the fall along with a research methods class, probability theory, linear models, and ODEs.
2) My GPA is pretty solid, so if I compliment that with an adequate\good GRE quant score I will apply to several top 20s.
3) If I apply to the top 20s, I will register for graduate analysis in the spring. The class has undergraduate RA has a prerequisite, but I should be able to obtain an override. I will take the class for credit/no credit.
4) The remainder of the schedule will consist of stat theory, advanced metrics, a thesis class, and PDEs or game theory.

The graduate analysis class will be tough, but I think I can scrape by with a 60%. If worse comes to worse, I can drop it. Signing up for it in the spring though accomplishes two goals. First, I develop my abstract math acumen and second it sends a strong signal to adcoms. One question: will taking the class pass/fail irritate the departments I am accepted to when they receive my transcript after graduation? When I apply my transcript will show the classes I have registered for in the spring, but not the grading option. Regardless, I feel like this is my best option. Thanks for the input.

Mankins
05-20-2009, 11:00 PM
Thanks for the responses. After taking what everyone said into consideration I've developed the following game plan:

1) I'll register for grad micro in the fall along with a research methods class, probability theory, linear models, and ODEs.
2) My GPA is pretty solid, so if I compliment that with an adequate\good GRE quant score I will apply to several top 20s.
3) If I apply to the top 20s, I will register for graduate analysis in the spring. The class has undergraduate RA has a prerequisite, but I should be able to obtain an override. I will take the class for credit/no credit.
4) The remainder of the schedule will consist of stat theory, advanced metrics, a thesis class, and PDEs or game theory.

The graduate analysis class will be tough, but I think I can scrape by with a 60%. If worse comes to worse, I can drop it. Signing up for it in the spring though accomplishes two goals. First, I develop my abstract math acumen and second it sends a strong signal to adcoms. One question: will taking the class pass/fail irritate the departments I am accepted to when they receive my transcript after graduation? When I apply my transcript will show the classes I have registered for in the spring, but not the grading option. Regardless, I feel like this is my best option. Thanks for the input.
The classes you "plan to take in the spring" don't send much of a signal to admissions committees. They won't see your grades and you can always drop the classes during the first week of the semester, so they have no reason to believe that they make you a more qualified candidate. I'm not saying you shouldn't take grad analysis in the spring, just don't do it because of its signaling value.