RWBG

My understanding is that agent-based models are a particular class of computation model that focus on programming simple rules and observing the complex emergent properties that develop from those simple rules. However (and here's where I start to move into territory I'm even less familiar with) I believe there are computational models that are basically using computers to program more complex (often dynamic) games in a more deductive fashion. The link you posted looks like an example of the latter (although even there, it looks like the computational model is ancilliary to the main part of the book, which looks like it consists of standard formal models). Edited for grammar and spelling.

I think you can feel pretty comfortable with NYU placing comparably with other top 15 schools; I think there's a chance it might place better. I really think it's a great place to be, but I do agree that (right now) it's hard to rank the training precisely. My guess is the reputation of the school will shoot up in the next few years, so I think you'll be very well-placed to enter the job market when you do.

I think you might be confusing heavily microfoundation dependent models (e.g. the dynamic stochastic general equilibrium (DSGE) models in macroeconomics) with agent based models, which don't have the same structure. Microfoundations are usually discussed when you have some larger level macro-phenomena (like how labor supply responds to an increase in aggregate demand) that you derive from rules about how individual agents in a system work, e.g. representative consumers, etc. - these kinds of models are not ABMs, as ABMs don't assume higher level structures, but just look at emergent properties. In fact, agent-based models have sometimes been suggested as alternatives to DSGE models, given that they make weaker assumptions about equilibrium, the behavior of agents, etc. Many ABMs don't even assume rationality, instead relying on evolutionary mechanisms, etc. But maybe you can give an example of work that uses an agent-based model in the way you're describing.

In my estimation, NYU's faculty well outstrips its reputation (and its reputation is quite good). I don't know how the training and placement are (and with a lot of the faculty being new, it's hard to get a sense right yet of how it'll do in the long term) but I think a good case could be made for going to NYU over any other school, at least if you're interested in formal/quant-heavy work in political economy.

At least to my knowledge, that's not actually how ABMs are used. The idea isn't to get a realistic account of the world, but to program a number of agents with simple rules and then examine what the emergent properties are of a system populated with those agents. For example: Axelrod's first computer tournament was an ABM-like system that didn't have a real-world analogue, but had numerous nontrivial implications when you looked at the results. ABMs also are interpreted way differently than formal theory models, in that formal theory models are more deductive, while ABMs give you a simulated data that can be examined inductively. It depends on what you're calling "economic tools". Certainly neo-Marxist work doesn't use tools being used by modern economists (those tools are basically all mathematical). From my experience, I'd say that the tools used by neo-Marxists are much more derived from sociology, although the work I've read definitely fits the category of looking at interaction between political and economic phenomena. Edit: I should note that I'm hardly that familiar with the neo-Marxist literature, so maybe you have examples of work that could be better characterized as using economic tools qualitatively?

Can you really look at political problems using economic tools qualitatively?

What do you mean by "a bit much"? Just curious!

Just removed a star from Yale in my signature! I'm enjoying the feeling of closure. Also, balledematch, I was under the impression that GT funded a fair number of their stronger applicants. I think you have a real shot there (in addition to Berkeley).

I agree with the two posters immediately before me. I will be turning down three of my four offers shortly after my last campus visit (i.e. some time near the end of March).

OH, I forgot, I'll set up one more mailing list; the MPSA mailing list. If you're going to MPSA and want to meet up for drinks (alcoholic or otherwise) PM me your e-mail.

I wrote a personal statement for Michigan, but it's silly enough that I don't really want to post it. Moreover, Otherworlder's post seems to suggest that it wasn't heavily weighted in the process (he can't remember writing it); I think it's a requirement of Rackham, but not a serious part of the admissions process.

I also think some of the responses to the initial post seem worth reading. Kristen Harkness' response seems good. I liked this part at the end: "We also care deeply about substantive issues and our research is driven by normative concerns over important global issues such as inequality, poverty, environmental degradation, social violence, government stability, and war. We believe, however, that good research design (whether quantitative or qualitative) is fundamental to drawing reliable inferences about the world in which we live. Without good methods, we cannot know the causes of the problems we study nor the consequences of the policy interventions we may design. And in the absence of that knowledge, we may do more harm than good."

Sigh... well I can't say this is a new criticism, although I continue to think it's mostly misplaced. Beyond that, I'll avoid commenting any further on this article. Edit: I will post this though. It's not about stats as much as it is about formal theory, but I still think it's a good read. http://www.jstor.org/stable/10.2307/2539250

I don't think it's good practice to soften the blow by attacking other schools. Everyone I know at Princeton seems nice.

So I'll be going to Madison, Rochester, UCLA, and Michigan, and I thought I'd set up mailing lists for those going to the weekends so we can meet up while we're there. If you're going and want to join a list, PM me your name and e-mail and I'll set things up. Your name and e-mail will only be shared with those also attending that school's weekend. Also, if you've been admitted to any of those schools but CAN'T go, you can ask to be added to the list so you can ask other attendees for information about the schools after they've been there/so you can discuss your thoughts about the school. Also, if you'll be visiting schools not on this list, I invite you to set up your own mailing lists using this thread!

Michigan actually sent rejections BEFORE admits. It was a little unsettling.

Official rejection from Princeton! Kind of a nice note, actually.

Congrats on the Yale admits!

Well, I don't know if CPE implies an emphasis on stats over formal. I think to some extent, how you rank the schools here depends on how much stats/formal you want. However, after thinking about it for a bit, I don't know if I could rank them for you. Schools like Princeton would be good if you want both the CPE substance and methods training, NYU's good if you want to do more formal stuff in addition to the stats, UCSD is good if your work has some cross-over with IPE (which much does). Admittedly, I'm much more familiar with IPE.

Congrats!!!

That's sounds unlikely (the increase in numbers). The past couple of years have already seen increased numbers due to weaker economic times. The only school I know precise numbers for (Michigan) had within 10 people of last year's number apply. What schools were you hearing this about?

I think Emory, WUSTL, and Vanderbilt are moving up. FSU's a bit more unstable; they hire a lot of good people, but a lot of the people leave soon after. It's overall I think very good, and it places well, but it's hard to pinpoint its direction. As far as Emory and more methods people, they did just hire Shawn Ramirez from Rochester, so maybe that's a start (although I think she might be more formal than methods)!

I'm a little unclear about the demarcation you're making between "econ-type (formal)" and the "cutting-edge methodology" of the "polisci type." Plenty of the political economists at Michigan are using formal models; did you just mean formal versus stats? But you do mean CPE and not IPE? If you can clarify what you mean, I can give my thoughts, whatever value they have coming from someone not yet in a Ph.D program.

I think it's a mistake to look at admissions decisions as having any level of consistency. It's not that you're less competitive relative to the other Africanists than last year, it's that when you have an entirely different group of people evaluating you, what's "competitive" changes significantly. It's quite possible that if you had applied with you application from today to last year's committee, you would have been admitted.

Don't forget, if your cycle ends, post your SOP! Help future applicants, especially those here who'll be back again next year; I know after last cycle, I wished more people had posted stuff.