Ondrej

Hi DemandSide, just a quick question related to the above - how do you deal with obvious fakery and people outright bulshitting in their SOPs? Or is it present in every application, so you just overlook it? Cheers.

Let's continue with a slightlyless plausible story, though still pointing to an important factor. Al and Lloyd have identical profiles, to the last detail. They have the same letters from the very same professors. They both apply to Awesome University. One day Al gets in a rejoices, Lloyd patiently awaits his result. Then something unexpected happens. They're out of blueberry muffins at the adcom's favourite coffee shop, she's rather disappointed for the day. She doesn't find Lloyds profile worthy of admission and dismisses his file. Regardless of Al's success the day before. You just never know. It's such a random process.

You make it sound so simple :) It's a 70, not really 70% (grading system is rather strange). Also you need a supervisor. And funding, which is hard to come by. So I wouldn't exactly call it a backup plan.

It's self selection as well, I don't see a lot of Cam people applying to the us. (Though there is one top 7 admit this year.)

I wouldn't read too much into it. They play the admissions game, they want the best students, naturally; but it's not like they're usually genuinely impressed by xyz's undergrad work. It's not like it's a job market thing when you've got (semi) serious research under your belt and you're getting a fulltime position. (Yes, I got an email like this, it was a nice touch, sure, but I wouldn't freak out about it.) Don't take it personally, my intention is not to offend. I agree with most of what was said above, I just don't think it should be a deciding factor or anything. Feel free to downvote. Cheers.

Um, some things I don't quite understand. 1) How come Cambridge is expensive, but Barcelona is not? The difference in fees is like 10% or so. 2) Good chances of getting into EME, but doubtful about Cambridge? That doesn't make sense to me. If you can afford it, go to LSE (and preferably take a gap year afterwards). You can also try to get into their PhD program. As far as I know LSE is top 10ish according to some rankings. Or did you mean strictly US top 10? Cheers.

If you're unsure about PhD, I'd go for UCL, because you get good connections in London, can go to workshops, interviews, job fairs, ... London is a great place to start a career. And if you wish to pursue a PhD, UCL is a great uni. If you do well, you get a decent chance of continuing there or getting to Oxbridge/Imperial/LSE/Warwick for PhD.

Depends greatly on what your plans are. In most cases I'd go for UCL, but I'll wait for your answer before giving my final opinion.

You mentioned lack of proxies and selection bias. One other major problem would be severe multicollinearity, especially in the top 15. Together with a relatively small sample it might be tricky to get anything out of the model.

I know of three people who got in just today (with full funding), but from what I gather, Mannheim is not their top choice. So it's not unlikely they'll reject this offer.

You can spell surprisingly well given your blood alcohol level. Oh and congrats.

Check out Goyal's work on networks.

Don't know that, but you can switch between the two. But mind that you can only "upgrade" to research if you do well in your math exam at the end of September.

I think so, both the uni choice and her last name would suggest that.

You might want to try contact this postgrad, she went to Utrecht for her undegrad and master's.