zzleeper

In stuff like financial economics - running regressions here and there - I believe most people use a mix of Matlab/Stata. Pure programming languages like C/Python may be important for specific applications, but there is always the chance that you may never end up using them. On a personal note, I believe that more important than the software per se is the community where it lives (if your prof uses it, if people in the topics you want to work on use it, etc.). Regarding the adcoms.. well, they have no way of knowing if you know ABCDX programming stuff, so I think they discount those skills quite heavily.

Well, this is probably my last post in this thread, but take it easy.. - Italics make stuff much harder to read, that's why they are not the default. - I asked you about the WB/IADB because I worked there in the past, in one case on-site and in another off-site (ie, not in the main building but in one of the several country offices). It's a small world you know.. - Replying to your questions costs us time and effort, which is scarce. Both capital_structure and I were trying to be friendly, and frankly, your replies are not very uplifting to say the least. - Many of us try to write well around here. English is not my native language and I tend to make many spelling/grammar mistakes, which is why I take my time to double-check my words before replying. You could do yourself a favor by doing that. Anyways, get your stuff together (doing some introspection is useful), and you'll have better chances with everything in life.

Good advice capital_structure.. By the way phd_world.. where did u worked on in the WB & IADB? in-site or off-site?

I just got admitted into a quite good finance program, and I learned a few things down the road. Let my try to share them with you: Be careful about teaching. Academia is NOT about teaching but about research. I also love to teach, but you kinda need to downplay that stuff in the SoPs. Pay a lot of attention to details. For instance, your SoP and CV need to be immaculate; and even stuff like your AWA matters. If I were you, I would retake my GRE and try to reach at least 780. Be realistic with your chances. Competition for top finance programs is very hard, specially if you don't come from the US. Aiming for some good European places may be a good way to hedge your bets. The usual question will arise: why are you trying to get into academia after four years of consulting? Specially in downturns, the adcom will suspect that its because of a difficult job market. They may also question your perseverance regarding the incomplete MBA. Regarding your SoP: Be frank. The adcom will spot any BS from a mile away. They won't be unimpressed by your knowledge of the M&M Theorem or of Basel II. Also, be aware that academic papers tend to be very dry and full of math, they are VERY different from what you read at an MBA (eg: Bodie-Kane-Marcus). When I did my masters in finance, a few friends with business/accounting undergrad got really burned out with the amount of math and technicalities. Anyways, good luck with your search; I just want you to be aware that the road ahead is hard and that you should be prepared. In any case, always consider other alternatives, such as masters program (specially in Europe as they are cheaper here. Eg: BGSE). Cheers, zz

I can only speak regarding finance, and not that much: - Getting into a top finance program is quite difficult to say the least, and a 750 GRE probably looks bad. - You forgot the most important variable, the letters of recommendation. Who are they? You need 2-3 professors to vouch enthusiastically for you. Even better, having LoR's from well known professors (at least to the AdCom.) is almost a must. - It's good to take math courses, but try to take something like graduate level real analysis or stochastic proc., specially for finance. - What are your research interests? What topic do you want to analyze and research about? Now, a honest question you need to ask yourself is why do you want a PhD. This question is REALLY important and many people fail to think it well. You are going to invest 5 years of your life doing research, learn new skills that are almost only applicable in academia, and even then, with an uphill battle for tenure. I know that my reply seems a bit harsh, but I'm just trying to be helpful (not trying to discourage you from pursuing a phd if that is what you really want). Best, zz

You are right fantiki, I've just accepted it. Congrats on UBC.

They care about the reputation of the program, measured through PhD placements. After putting a lot of effort into its MSc programs in recent years (through the creation of the BGSE, increased funding, etc.), they now have a steady supply of good 1st year students. The next step is to improve the job market placements of the PhD graduates. This means that they would STRONGLY prefer to keep their 2nd year PhD students, which have already proven their analytic skills, and are almost midway through the PhD program. The problem is that many students don't want to stay, specially those more academically inclined / with better grades. There are many good things about UPF, specially their very strong macro/international faculty (through CREI), but one of their weakest points is the (perceived?) lack of approachability of its faculty. Since most of the research is theoretical, there is almost nil collaboration with students. I also feel that the faculty commitment towards the success of PhD students is weak. Take these last points with a grain of salt, however, since I may be wrong, and in any case I'm just generalizing.

I arrived with the impression that most people used UPF as a bridge to a top US school; that was my intention and I thought the faculty were Ok with that. When asked by the department chair, I answered honestly about that. I've now become less naive. I doubt you would "burn bridges", it's just than you should not expect much support through letters or advice, specially in with senior faculty (they are more interested in the success of the program and not of the students per se)

I agree with invisiblefist. Having a VERY STRONG math background is a must; having some RA experience and maybe publications (which I lacked) also helps a lot. Finally, I think LoRs matter much more than you think.

If you want to get into a good phd program, it would be wise to get into an international (Europe/US) program to get good LoR's, etc. Otherwise, programs would be very reluctant to take such a huge risk with unknown universities / professors.

I don't know much about Tinbergen but it seems like a decent place. Check the details about the faculty (specially those that will teach you) to get a better idea. I did the msc in barcelona. If you have any particular questions, feel free to ask. In general, I was lucky to have some very good teachers (Arturo Bris, Tomas Bjork, etc.), although they are not UPF faculty (eg: Bjork is from Stokholm). The finance faculty at UPF is really small, which is a problem if you think about staying for the PhD (except if you are interested in banking).

Ok, so to answer your profile. - Be incredibly careful with all your profile: when everyone has 800q , a 500v in GRE may hurt you (same for essay) - Does your RA experience has something to do with finance? - How well known are your LoR's to the AdCom? Is there a way for you to get better LoRs? (through this year?) - Is there any math course (real analysis, stoch proc) that you could take next term? Good luck

Is there any reason for two "deleted posts"?

I think no news is bad news. I don't know for sure but I believe they are done with their interviews.

Sorry for the late reply. I received it. I believe the acceptances are over, but I'm not sure about waitlists.

Sure: Downloads for sergiocorreia's GoogleMatch - GitHub That said, after rereading it, I would think that you would need someone with some basic knowledge of python to fix the script for your needs (like a friend from the CS department). For instance, I went for the first .EDU , not just the first result, since I was working for universities. I also did several other tweaks, like implementing some cache (if the connection stopped at 7999 obs, it would resume from there), implement a blacklist (eg: no wikipedia results), and manually correct some cases. As with anything in life, the first 90% of the program is quite simple and took just a couple of hours. Tweaking it, on the other hand, took me the rest of that day =/

I had to do similar matches to join three datasets containing university names. There were around 8000 obs also, and I would have probably resigned before doing that by hand. In the end, I did this: I made a python script that googles "Larsen & Toubro" and "L & T Ltd." As you see.. they give you the same url ( http://www.larsentoubro.com ) , which you can then use as the link to do the join. L & T Ltd - Google Search. Larsen & Toubro - Google Search I thought it was pretty neat, although the prof only cared about the results being ok, not the neatness of the process :(

Don't use SAS; it sucks monkey balls, it's syntax is stupid, and it's extremely expensive. If the data is not THAT big (ie, more than a few MILLION! observations), use Stata/Excel+VBA or whatever you want except SAS. I had to do something very similar a little while. For the date variables; come on.. there are at MOST a couple possible ways of reporting dates. In my case, Stata has very flexible commands for entering dates from strings, so I used that. For the names, you would be better off trying to do a fuzzy join. For instance, how could you join "Siemens" and "Siemens Ltd"? A few ways are: 1. Quickly remove the Ltd. and the trailing spaces from the names. 2. You could apply a "diff" algorithm to see how many changes are required to go from one name to the other. In our example, its 4 changes. 3. There are a few more options but increasingly more complex, as matching by url (make a program that does a google search and get the url), matching by NYSE ticker (similar), etc. You can try any of this, but for God's sake, plz dont do it by hand!! Good luck!

How much space was the data occupying? (after compressing). Maybe you can do some rsync every couple of days, it shouldn't be that expensive. Hard drive failure should be thought as a normal occurrence, which everyone should plan for. BTW, this thing happened a few weeks ago at StackOverflow (a very popular forum for programmers). You may be interested in the solutions they considered: Blog Outage – Backup Policies - Blog – Stack Overflow

Where do you want to go later? Top10 USA? If you wait one or two weeks, I may be able to tell you how well (or bad) we did this year. That said, always think carefully about your choice of LoRs. Don't just go for the big names if you are not 101% sure about their enthusiastic support. Can you explain more that part about leaving after M2? Also, last year there were very good placements from 1st year BGSE students. Two guys to Chicago, one to NYU, and one to MIT (although the last was previously an undergrad UPF student). That said, they applied just after arriving to BGSE, so most had LoRs from somewhere else.

That would be great to know =)

AFAIK, they are holding phone interviews this week.