thagzone

You don't mean every school in OP's list, do you? BU, Boston College, WUSTL, and UIUC (hell maybe even Georgetown) may not get you tenure track jobs at a T50 with any regularity, but I wouldn't say the employment options coming from these programs are terrible. I may come closer to agreeing on the rest of the list.

Susan Athey is a good start. OP, what exactly are you interested in with regards to ML in economics? You won't find economists developing new ML algorithms, but you will find a variety of applications. (1) Some such as Chernozhukov at MIT are interested in areas like dimensionality reduction and using ML to estimate nuisance parameters. There is an area which probably begins with Chernozhukov at the top interns of rigor, and which focuses on deriving traditional econometric properties of estimators that are fully, or augmented by, ML-based. Simulation and performance are also considered. (2) There is another area of researchers that are simply interested in what it would be like to replace traditional econometric estimation with ML. This is less likely to get you very far for a lot of reasons (chiefly: causality still rules the day in academic economics). See, for instance, Bajari et al. throwing 8 different ML methods at a traditional demand estimation problem and benchmarking out-of-sample performance against traditional IO models (there is some straw-manning happening here, but it's not that bad). (3) Finally, there is another group of empirical researchers who don't "do ML" per se, but are interested in using ML for creative purposes such as unlocking new data sets, augmenting additional data sets, or benchmarking their econometric models in terms of out-of-sample predictive performance. No one here is involving ML at all in their statistical results of interest, they're just using it as a step in the process before the real econometrics are done. These are, at least, the main categorizations as I see them right now in academia. I see (3) as being the most accessible and the easiest to get a foothold in, (1) as the most challenging and the least likely to be feasible unless you're in a top econometrics department, and (2) to be worth our while as a field to take a stab it--if nothing else out of curiosity, but very crude at this point and unlikely to look good as the core of a dissertation or JMP. Economists working at places like Google and Amazon will be stepping even further outside of these boundaries. I'm personally working on combining aspects of (1) and (3) above. I'd be happy to point you to more resources if the area interests you. I'd say you need a strong foundation in econometrics, as well as coding chops and some computer science background. Beyond that, outside of the top 5, working in ML will require you to be creative with the resources at your disposal and you'll have to be willing to go outside of your own department frequently.

Maryland insider here. There have not been decisions yet, and I don't even think the admissions committee has begun meeting. As said above, it'll be clear on GC when Maryland's offers are actually released.

Submit the first one. Especially if you are not a native speaker. 3.5AWA isn't great, but 161Q is a much bigger red flag in economics. Depending on the masters programs you're applying to, it may not matter. That said, choose the first option.

Really? Then you ought to keep reading. But in addition, a lot of people who have thought deeply enough about a PhD in economics to make it to this forum already have pretty well calibrated their expectations. Most people are targeting a reasonable range of schools when they decide to post here for advice by selection. The ones that aren't get a little nudge in the right direction. Occasionally some posters have no idea what's going on, and they are set straight pretty quickly. In general, members on this forum with some experience are not trying to make people feel good. We've applied, we've spent loads of time and money, and we know what a trying and expensive endeavor this is. If you were overshooting, I'd tell you, and I'm pretty sure almost everyone else here would too. Within the last couple of weeks I recall specifically telling a prospective applicant they needed to add more schools further down their list or they were risking getting shut out. If I just wanted to make people feel good, I'd stand around campus with one of those "free hugs" signs. In your case, you've really got a pretty good looking profile with a couple of caveats--one big and one small. The small one is that your overall GPA is a bit low for a successful T20 applicant, but your stellar econ and math grades should really smooth that over quite nicely. The large one is that you seem to have little research experience to speak of (unless you do and just omitted it for some reason). The lack of research experience means that your range of outcomes is quite large right now. I could see you getting into a borderline T10 school if one of your letter writers pulled strings for you, and I could also see you not getting much outside of the 25-40 range. On the other hand, if you spent 1-2 years at a good full-time RA gig, I am highly confident you'd get a few T20 offers and feel there's a decent chance you'd get some T10 offers. My analyses are only ever objective--do with it what you will.

It's not that it won't harm you at all. Even if you get an A in real analysis, certainly, on the margin, it could still be a detriment. It's much less likely to negatively impact your application outcomes if you do get an A in more advanced proof-writing classes. As cyjNel says, though, you'd better figure out what went wrong in your Intro to Proof class, or getting an A in real analysis is a pipe dream anyway. Get the textbook that will be used before the semester begins (hopefully it's Baby Rudin) and start working through the material on your own.

This is perfect advice. Your profile is actually sort of unique in that its weakness is directly identifiable and completely within your control at this point. With 1-2 years in a high-level full-time RA position (you should be a good candidate for these), you could likely crack the top 10.

I think your range is not bad. I would expect you to at least get a few admits in the 10-20 range.

Both, but more so your lack of research experience. Your math background matters more if you are intent on pitching yourself as a theory student. I don't mean you shouldn't try T10-T15, but if you aren't financially constrained, you really should put out some more applications in the 20-40 region.

In my opinion, you are overshooting. You should be applying to more schools in the 20-40 range.

Higher Q only, without a doubt.

I wouldn't do that if I were you. It's going to be a nuisance for the adcoms, they're probably not going to read it, and it'll obscure the important stuff that they do want to read. What to kick out, then? Nothing I guess, just be more concise. It can be done.

Any standard calculus book that covers through multivariable calculus should be suitable. Really though, the majority of calculus should be a reflex by the time you start a PhD--if you didn't learn it very well during your university courses, then that may be a sign of deeper problems.

Not noticeably. But that's meaningless--very small sample size. The Americans without a masters that I know in my program and in others generally spent a year or two or possibly more doing one of the things you mention after undergrad. You'll find that the backgrounds of your cohort mates are far more heterogeneous than you might believe from perusing application advice threads and blogs.

Exactly this. I'm also not a first-year, but the 5 American students in my cohort didn't have a master's degree, while 11 of the 12 foreign students did. I believe this is pretty much the norm.

Yes, agreed. I was kind of surprised to read OP say that. To add another, Maryland's AREC program doesn't require a masters. Your research experience is rather sparse, but you otherwise seem to have a fairly competitive profile for direct entry into an ag/applied econ PhD program. What makes you think you're not ready?

Funny, when I read this earlier but didn't have time to comment at the moment, this was my near-verbatim reaction.

They will likely only look at writing samples in borderline cases. In general, these will not make much of a difference in your admissions outcome. As far as your question about the composition of the admissions committee: I find it highly doubtful that something written by an undergraduate or masters students is so field-specific and esoteric that it would take a person exclusively in that field to appreciate it as being either good/indicative of highly developed research potential or unimpressive nonsense. If it's a good writing sample, and an admissions committee does look at it, they'll know it is.

Assuming you've got a good host of undergraduate economics courses from a U.S. university, a Masters program may not be all that useful. Without a complete profile it's hard to say, but given that you're late to the game, you may lack research experience. Masters programs can be quite expensive, so if money is any kind of issue for you, consider looking for RA jobs and taking a few courses as a non-degree student to round out your profile.

I would agree with that. An introduction to econometrics is very important to have as well, of course. A first-year PhD sequence will essentially teach econometrics from scratch, but it will be highly theoretical, abstract, and will move extremely quickly. While it is possible for an otherwise prepared student with no econometrics background to keep up, it is my impression that it would have been extremely difficult for me to get through first-year econometrics without my undergraduate coursework in it. As others have said, delaying your application for a year to round out some of these other deficiencies would be prudent.

You are correct about that. In addition, you're underestimating the importance of proof-writing skills. Definitely heed the advice already given and take the abstract math course.

Oh my, what a terrible spot to be in! But really, I guess you'd take the perfect quant score, since a 168 -- > 170 verbal difference is immaterial anyway and a 5.5 --> 5.0 AW difference is also never going to matter. A 168 --> 170 quant score increase is not likely going to change anything itself, but I guess you could argue it's the most likely to help you, while the other two shouldn't make a bit of difference either way.

I'll let others with more experience speculate on your floor and ceiling, but my gut is that, yes, schools will be willing to mostly overlook your first attempt at college given how incredibly different your second go at it was. Two quick things that stand out: swap differential equations for real analysis if you can, and retake the GRE. The Q score can be over-emphasized at times (e.g. bickering about whether or not someone with a 165 really needs a 167+), but a 159 is low enough that it may really do damage to your chances. What is your research background like and what sort of letters or rec will you be able to get? The answers to those questions will largely determine what your prospects are.

Whether or not a semester abroad will have a detrimental effect on your profile depends on where you stand right now. If you have a lot of ground to cover, it may impact you adversely. If you are already have built a decent profile, it likely won't matter at all. In general, I am hugely in favor of choosing to spend a semester abroad during your undergraduate years if you have the option--likely for the same non-academic benefits you allude to.

I would agree here. If you're doing too much taxing simulation/empirical work on your personal laptop, you're probably doing it wrong. Learn how to work a little in Linux and then you'll never have a reason not to use your university's computing cluster. As far as laptops go, you probably don't want less than 8gb ram these days in general, and if you're getting a new machine, you should absolutely be making sure to get a solid state hard drive.