Hi,I'm thinking of starting a PhD very soon. I've just been offered a choice between doing pure theory (theoretical physics) and computation (computational physics). The problem is I actually would rather do computational physics but my fear is that in three years time when I take my PhD to the jobs market, it and therefore I, will be considered second rate next to those candidates who have a PhD in pure theory. I would be very grateful for any advice offered. Thanks. PS - my concern is the result of a rather dismissive comment made by an academic I recently spoke with, when I told him about the choice I had to make he referred to theory being much better than "only more numerical projects". I personally hope that snobbery at this level does not exist among quants.

I still have a hard time understanding why some people go for a Physics PhD when they know they want to work in finance. It's fine to do a PhD in Physics and then realize you want to work in finance. But if you already know you want to work in Finance, why don't you do a PhD in Finance or Applied Maths?

I heard a talk by Emanuel Derman a few months ago where he discussed this point and asserted that doing a physics doctorate as a way of becoming a quant was just dumb (today) now that Quant Finance / Financial Engineering has a healthy theoretical basis. Why not do a PhD at Columbia, NYU, CMU, LSE, Birkbeck, or a French university?

QuoteOriginally posted by: mattdamonHi,PS - my concern is the result of a rather dismissive comment made by an academic I recently spoke with, when I told him about the choice I had to make he referred to theory being much better than "only more numerical projects". I personally hope that snobbery at this level does not exist among quants.I personally hope that undergrads would not call academics snobs after asking for an advice.

Sorry, let me clear up the confusion I've created - I've chosen to go for a physics PhD not because I want to be a quant but because I do actually want to do a physics PhD. I posted the thread simply because I happen to see quant as being one of a few options which down the road I may wish to explore. I was a bit concerned by this guy's remarks so I thought I'd see what you all have to say and whether or not you agree with his point of view. As it happens one or two people on this forum have already been very helpful in being kind enough to give me their opinions, which I take very seriously. I just thought I'd broadcast to the masses in order to get some kind of consensus view I suppose.Incidentally, I don't believe Emanuel Derman said quite what one of you claim he said, something similar, yes, but not quite the same thing.Neige, I can see where you're coming from, but I should have been clearer. The fact is, I didn't actually ask for his "advice" in the first place, and nor did I get it as far as I can tell. What I got instead could not by anyone be misconstrued as advice. He was making what seemed to be a disparaging remark about the nature of numerical work which as it happens a good proportion of his colleagues spend most of their days sweating over. If that's not snobbery, what exactly is it? I am by no means disrespectful and frequently can be seen addressing academics by their formal titles and leaving apples on their desks.Finally, it was just a question - does such academic snobbery exist among quants, nothing more, not an opinion.

alohashirt and valderama make a valid point. And young man, you need to decide what you want to do when you grow up.And one final remark:quants, as I'm sure you're aware, spend > 2/3 of their time coding ( the proverbial computer monkey so to speak ), so "academic snobbery" aside, you'd be on the computational side of things if you become a quant.

QuoteOriginally posted by: mattdamonSorry, let me clear up the confusion I've created - I've chosen to go for a physics PhD not because I want to be a quant but because I do actually want to do a physics PhD. I posted the thread simply because I happen to see quant as being one of a few options which down the road I may wish to explore. I was a bit concerned by this guy's remarks so I thought I'd see what you all have to say and whether or not you agree with his point of view. As it happens one or two people on this forum have already been very helpful in being kind enough to give me their opinions, which I take very seriously. I just thought I'd broadcast to the masses in order to get some kind of consensus view I suppose.Incidentally, I don't believe Emanuel Derman said quite what one of you claim he said, something similar, yes, but not quite the same thing.Neige, I can see where you're coming from, but I should have been clearer. The fact is, I didn't actually ask for his "advice" in the first place, and nor did I get it as far as I can tell. What I got instead could not by anyone be misconstrued as advice. He was making what seemed to be a disparaging remark about the nature of numerical work which as it happens a good proportion of his colleagues spend most of their days sweating over. If that's not snobbery, what exactly is it? I am by no means disrespectful and frequently can be seen addressing academics by their formal titles and leaving apples on their desks.Finally, it was just a question - does such academic snobbery exist among quants, nothing more, not an opinion.If you don't want to be a physics professor or work as researcher at Intel or Microsoft, just don't do a Physics PhD. Just do a PhD in something else, or better, do just a master's degree and work directly after, or don't even do a master's and work directly. Again, PhDs are for people who want to become professors or researchers, especially in Physics.

Thanks Valderama for that advice, but to be honest right now I can't think of anything better to do than research for three years. Surely three years isn't so bad, I quite like the student lifestyle anyway. I've spoken to quite a few people who have either completed a PhD or are in the process of completing one, and you'd be surprised how many of them say that the reason they undertook it in the first place was because they couldn't think of anything better to do. Personally, I believe that so long as I work on something interesting, get along with my supervisor, and have time off (maybe not much) to relax and do something completely different, then provided I complete it and get the qualification, I won't regret it, I would consider it an accomplishment in itself and just move on.

phds are a long hard slog and the post-grad lifestyle is nothing like the undergrad oneYou should only do a phd if you are so into the subject that you believe that you will spend the rest of your life regretting it if you don't do one.

.....help?

Hi mattdamon,Your question is fundamental. Basically what I think you are asking is whether or not you should follow your desires, and whether these desires are somehow impractical in such a way as to limit your options in the future. Specifically, you want to know if there is a derogatory attitude in quant-land toward computational physics relative to theoretical physics. The answer is that sure, there probably are some people that have such an attitude, but I believe this number would be small compared to the number of people that would regard computational skill as a real plus. Moreover, quants are a pretty diverse lot. I know quants who have degrees in various kinds of physics, both experimental and theoretical, including elementary particles, biophysics, geophysics, and general relativity. There are quants with degrees in engineering of all sorts (chemical, mechanical, electrical, nuclear, fluid dynamics), computer science, pure math, etc - you name it. Besides, there are many variables besides your background that influence whether or not you get a job in any given field, including finance. Even at a University, getting a job depends on your production and reputation, the fit within the department, the number and ability of people competing with you, etc.Having said all that, I think people on the Forum so far have given some pretty good advice to you. My advice would be to do what you really want to do and forget about academic snobbery. I would seriously advise that if you do computational physics, try to obtain as many practical mathematical skills as possible. On the other hand, theoretical physics these days can have more numerical aspects than you might think. In any case, try to come out with a well-rounded set of skills and knowledge. Bon voyage. ---------

Hi MattDamon, Do exactly what your heart desires when it comes to academic studies. You will only regret it if you follow someone else's advice and your heart really isn't in it. Go with your (unique) strengths and God-given talents and you can't go wrong. Going for the money, when it's not something you enjoy, doesn't work. I have learnt both of the above from my own experience. Yes, it was painful. Oh, and Good Will Hunting wasn't such a bad film either .TJ.

Besides, there are snobs everywhere. You can't let them influence the path you choose. I'm sure some theoretical physicists working on a particular problem deride those working in a different area of theoretical physics. As for quants, I came across one who develops complex models to price complex instruments, saying derogatory things about certain simpler models someone else was developing. This, half an hour after being unable to explain why the market prices were so widely off from his model predictions and implicitly admitting that the models were for all practical purposes, useless! Does the emperor have no clothes?

Singlestrand: Besides, there are snobs everywhere. You can't let them influence the path you choose. I'm sure some theoretical physicists working on a particular problem deride those working in a different area of theoretical physics.Yes, check phenomenologists (e.g., mikebell) vs string theorists (Michio Kaku), for example. The debate can get quite quickly become heated and intense to say the least. My old professors (strict phenomenologists) vehemently opposed anyone doing string theory, mainly because their own mathemetical capabalities limited them and to a lesser extent (their excuse) string theory lead to no experimentally verifiable predictions at that moment.

Fast forward 3 years.You're in the Royal Exchange sipping a gratuitously overpriced coffee with Paul & me.You talk of your PhD. You got there in the end, but there seems little enthusiasm in your work. You have some vague knowledge of the programming so many of our clients, but consists of impressionist Java and if you were naive enough to listen to an advisor at Imperial, Delphi. You did something quite hard, but it wasn't the thing you were best at so isn't the most outstanding lump of research.Alternatively, you sparke with enthusiasm about some advanced algorithms you devised. You have mastered C++, hell maybe you simulated fluid flows on a GPU, and we are very impressed. Guess which version of you gets forwarded to some big bank ?I don't believe any HH will prefer "pure" physics over computational. I cannot claim to have a full sample set, but my guess is 50% of managers will actually prefer it, and >25% of the rest won't care much either way.This of course depends on the detail of the two PhD's. Some bits of "pure" are really applicable, and some bits of "computational" may be stuff like symbolic computation of large analytical solutions which is hard but currently of zero relevance.For a headhunter good PhD contains the following attributesFrom somewhere good.Math based, pref some sort of PhysicsIt must be "hard". Define this as at least one word in the title we have to go and look up.Not too arcance, Define as three words we have to look upSome programming skillsA recognizable achievementGood programming skills, but not "operations", ie housekeeping type crap.