I'm told that theoretical physicist can go into investment banking

[twofish-quant]

Depending on the state of the economy when you apply, you may only get one or two interviews from different firms.

That's not usually the situation. Most large firms are confederations of different groups, and it's not uncommon in some firms to be interviewed by two different groups at the same time. Also, if you blow one interview, there is usually a six months waiting period before you get another shot.

There are about ten or so large investment banks, but they each have dozens of groups internally. There are *thousands* of hedge funds.

It happens that with my current employer, I got my job after my third try (with a gap of one or two years between interviews).

Therefore it's crucial that you prepare well, as you may not get the opportunity to use the first few interviews you get as mere "practice".

The problem is that your first few interviews are going to be practice anyway. No matter how hard or how much you prepare, there is a big difference between reading about riding a bicycle and doing it, and you should expect your first two or three interviews to go badly. It's part of your education.

Also for Ph.D.'s, the reason what finance is attractive is that the jobs exist. If you have to go through insane hoops to get a job in finance, than you are better off looking elsewhere.

 

[atyy]

Quantum finance: path integrals and Hamiltonians for options and interest rates by Baaquie

Path integrals in quantum mechanics, statistics, polymer physics, and financial markets by Kleinert

Physics of finance: gauge modelling in non-equilibrium pricing by Illinski

Also Illinski's book is worth getting even if you aren't going finance, because the first few chapters have the most readable explanations of fiber bundles that I've ever seen.

How interesting to see these recommended by someone who works in finance! I came across the first two when trying to learn stochastic processes for biology, and wondered if real finance people use this. Illinski's title sounds interesting, I'll definitely take a look.

 

[twofish-quant]

I came across the first two when trying to learn stochastic processes for biology, and wondered if real finance people use this.

It's complicated, because what types of math gets used changes from year to year. One thing that is educational is to compare Baaquie's 2004 book with his 2009 book. The math techniques that were used in 2009 are somewhat different than in 2004. One suggestion that I have is to read these sorts of things as "history books." Reading a text from 2004 will give you an idea of the math techniques that were used in 2002, which may or may not be related to anything anyone is doing in 2011.

One reason that finance books have surprisingly readable explanations of math is that traders will not use (nor should they use) math that they do not understand, so there is a lot of pressure to try to explain what you are trying to do.

In the last few years, monte carlo has been really popular, because

1) computers are a lot faster in 2009 than they were in 2002
2) people are much more suspicious of complex equations in 2009 than they were in 2002. One good thing about monte carlo is that it's easy to explain.

 

[atyy]

It's complicated, because what types of math gets used changes from year to year. One thing that is educational is to compare Baaquie's 2004 book with his 2009 book. The math techniques that were used in 2009 are somewhat different than in 2004. One suggestion that I have is to read these sorts of things as "history books." Reading a text from 2004 will give you an idea of the math techniques that were used in 2002, which may or may not be related to anything anyone is doing in 2011.

One reason that finance books have surprisingly readable explanations of math is that traders will not use (nor should they use) math that they do not understand, so there is a lot of pressure to try to explain what you are trying to do.

In the last few years, monte carlo has been really popular, because

1) computers are a lot faster in 2009 than they were in 2002
2) people are much more suspicious of complex equations in 2009 than they were in 2002. One good thing about monte carlo is that it's easy to explain.

In science, there are also experimentalists and phenomenologists - the books you recommend seem more theory based? Are there books for finance "experimentalists" or "phenomenologists". Also, is there any kind of "industry standard" process by which data is gathered and a theory is verified in finance?

 

[twofish-quant]

In science, there are also experimentalists and phenomenologists - the books you recommend seem more theory based?

That's because I'm a theorist. There are people with experimental backgrounds in finance.

Are there books for finance "experimentalists" or "phenomenologists".

It's quite a bit harder to find those books for cultural reasons. Basically, theorists tend to work on complex derivative valuation, and in that area there is pressure to be more open about what you are working on, because you are making money selling derivatives and if you have a model that everyone knows, you are more likely to be able to make the sale. There's another marketing reason in that there are people in the industry who are "stars" (i.e. everyone knows who they are) and as with academia, one way of be a star is to write a book.

Another game is algorithmic trading. What you do is data mining and time series analysis, and this is one area that is strong for people with statistical and experimental backgrounds. It's also a lot more secretive. If you figure out that X, Y, and Z are buy and sell signals, then the last thing you want to do is to write a book on it. The good terms "market microstructure" and "quantitative trading" will get you some books on the topic.

There are also some regulatory issues. It's easier to talk about derivative pricing because in the United States, most people cannot buy derivatives. If I start talking about what stocks to buy or not to buy, then SEC is going to get very, very annoyed.

Also, is there any kind of "industry standard" process by which data is gathered and a theory is verified in finance?

You look at your bonus check at the end of the year.

One thing that I do like about Wall Street is how everything is focused on profit. This does have some bad effects, but it also has some good things, in that you tend not to get into endless random discussions that go nowhere.

There are some interesting philosophical issues here. One problem is that the rules change. For example, if you did a medical study in 2005 or did some physics experiments, the one thing that you know is that the laws of physics and biology didn't change much if at all between 2005 and 2011. However, one thing that we do know is that the rules of finance changed a lot between August 2007 and December 2007.

The other thing is that there is something I call "physics-truth" (The speed of light is constant). There is also something I call "marketing-truth" (Coke tastes great). The weird thing about finance and economics which I find fascinating is it doesn't fit into these two categories.

 

[DrummingAtom]

Also, is there any kind of "industry standard" process by which data is gathered and a theory is verified in finance?

You look at your bonus check at the end of the year.

I don't know why but this is one of the funniest things I've ever read on here even though I know you're being serious.

 

[ode_to_joy]

Twofish-Quant, would you advise undergraduate students to take few economic/financial electives if they are serious about this path? I mean, is it really necessary to take these courses if you want to get a job in finance?

 

[jetwaterluffy]

You look at your bonus check at the end of the year.

One thing that I do like about Wall Street is how everything is focused on profit. This does have some bad effects, but it also has some good things, in that you tend not to get into endless random discussions that go nowhere.

If finance was scientific, they wouldn't have bonuses, as they stress people out so much it actually lowers their productivity.

 

[twofish-quant]

Twofish-Quant, would you advise undergraduate students to take few economic/financial electives if they are serious about this path?

You can save yourself some money, and buy some books on Amazon. It's important to get a strong undergraduate liberal arts education, so you might be better off taking a course on 19th century French literature and buying economics books on Amazon, or taking a course on economics and buying 19th century French literature. Or learn sculpture, or philosophy.

One thing to keep in mind with economics and finance is that when you read a book on Newtonian mechanics or algebraic topology, you can be pretty sure that what you are reading is not false. This isn't true with economics and finance. Much of what you read in an economics and finance text are flat out wrong. The hard part is that no one (including myself or the professor) are sure which parts are wrong.

One thing that I like about finance is that *everything* is relevant. I know someone that had to teach himself about the industrial uses of soda ash because he had to figure out how much to loan to soda ash companies. There are people I know in finance that come from art backgrounds, because when someone wants to use a painting as collateral, you have to know how much it is worth.

So for undergraduate it's less important to learn special topics, than to "learn how to learn."

I mean, is it really necessary to take these courses if you want to get a job in finance?

It's not essential. I've never taken a formal course in finance or economics. I have a ton of finance and economics books on my bookshelf. One reason this works is that I happen to find the topic interesting.

 

[ode_to_joy]

I want to get a job in quantitative finance right after my phD. I want to earn $200k+ per year and I want to get away from publish or perish pressure. You have mentioned that the subfield of mathematics (ie number theory or topology) does not really matter. How about abstract algebra? I believe groups, fields and that kind of stuff or logic hardly cover quantitative method in their curricula, but is it still possible to get a quant job with phd in set theory, logic, or abstract algebra? I hope this relativelt metamathematical field somehow provide a different perspective on finance but I am not sure if this is the way it works

 

[Vanadium 50]

8 days ago you were dhttps://www.physicsforums.com/showthread.php?p=3560353#post3560353" The post-PhD time scale is 350 or 400 times longer than this. So it's important not to get ahead of yourself - doing well in high school is your next step.

 

[Raioneru]

One of the paradoxes is that if you want to get a finance related job through the Ph.D. physics route (and this is a bad idea unless you are interested in physics), you are better off focusing on doing research on something in physics than in something financial. The typical research topics that theoretical physicists work on turn out to be more relevant to certain financial jobs than the stuff that people in finance work on (which is why banks hire physicists).

It's also extremely difficult to do research on financial topics part time, because in most situations, you are not going to have access to the raw data. One of the differences between finance and physics is that if you discover some deep secret about how black holes work, you want everyone to know, whereas if you discover some deep secret about how the stock market works, then you want no one else to know.

Basically I have 2 plans of equal importance (I guess I'll just go with what opportunity I'm given). I would like if possible a diverse career with options. I'm pretty interested by the application of physicist skills to the financial market but I would also like to have the possibility to work as a university professor, I really want to at least teach physics at some point in my life, its a big deal to me. It also explains why I don't want to get a MSFE after all: It would limit my option and inevitably make me regret deviating from my physicist path.
(comparatively Medical Physics would give me the same options: R&D/teaching/clinical practice/administration etc. . .)

Seriously twofish-quant, I would like to thank you for sharing your thoughts and knowledge about the subject. you are extremely helpful :)

 

[twofish-quant]

Basically I have 2 plans of equal importance (I guess I'll just go with what opportunity I'm given).

First you have to decide what you really want to do with your life.

In my case, my primary goal is do "do physics" and if some major research university offered me a tenured faculty positions paying one third of what I make right now, I'd be gone in an instant.

As it is, I haven't gotten any job offers, so I'm going to have to make do with what I'm doing.

The reason that this matters is that if you have any intrinsic interest in finance, then getting a physics Ph.D. makes no sense at all. Getting a physics Ph.D. is probably the worst way of doing finance if that's your main interest. However, my main interest is doing physics, and what I'm doing is as close as I can get.

This also matters. Right now about 50% of my job involves administrivia, and only about 25% involves anything like "research." That's bearable, but if my boss comes in and I figure out that my job has been changed so that I'm not doing anything geeky at all, I'm also leaving for something else.

I would like if possible a diverse career with options. I'm pretty interested by the application of physicist skills to the financial market

One problem is that if you are starting your physics Ph.D. now, then the world is going to be a very different place in a decade.

but I would also like to have the possibility to work as a university professor, I really want to at least teach physics at some point in my life, its a big deal to me.

Be careful what you wish for.

If you want to teach physics, that's easy. Community colleges are looking for adjunct physics instructors, and you'll be working as a TA as a graduate student. Now, if one thing that you'll learn about teaching is that it's hard and sometimes rather annoying.

Seriously twofish-quant, I would like to thank you for sharing your thoughts and knowledge about the subject. you are extremely helpful :)

First figure out what it is that you really want to do. Then figure out what you can get given that you can't get what you want.

 

[icomeinpeace]

twofish-quant,

What would be your advice for a high-school student who wants to become a quant? I thought a sequence like the following was perfect for that goal:

Undergraduate:
http://ugradcalendar.uwaterloo.ca/page/MATH-Actuarial-Science-Mathematical-Finance

and then any of the well recognized MFEs on the market. But then I read this thread and your opinions against MFEs, and going into something like Physics if the ultimate goal is to become a quant. I agree with the latter, but if MFEs are not a good start for quant jobs either, how else can anyone get into the field?

Thanks a lot...

 

[twofish-quant]

What would be your advice for a high-school student who wants to become a quant?

Get a well rounded liberal arts education in college, try different things, and don't be too quick to decide on what you want to do.

Realize that by the time you are 30, quants may not exist, and the hot job may be something in repairing solar panels or in agriculture.

Also, since you are in Toronto, it could be useful to take part in the protests that people are setting up. Lots of useful experience and you get to meet people.

I thought a sequence like the following was perfect for that goal

Don't think so. I don't see any art, history, philosophy, or literature in that program. Also it's important in undergraduate years to be different. If you take the exactly same courses and do the exact some coursework as everyone else, then there is nothing that makes you different when employers look for people, and you are going to be one of a ton of people after the same set of jobs.

If you study goat-herding in Tajikistan, then there is something different about you which can be useful if there is a job that involves Tajik goat herding. (It turns out that one of the major financial writers started studying Tajik goat herders. It was useful because when the Soviet Union fell, they needed a Russian speaking reporter.)

But then I read this thread and your opinions against MFEs, and going into something like Physics if the ultimate goal is to become a quant.

My ultimate goal is and also has been to study astrophysics. It turns out that being a quant is the closest thing that will get me to that goal.

If your ultimate goal is to be a quant, the first thing that you should ask yourself is "why do you want to be a quant?" In my case, it's because it's the closest thing to computational astrophysics that I can find, and if I can find anything closer, I'll quit.

I agree with the latter, but if MFEs are not a good start for quant jobs either, how else can anyone get into the field?

No clue at all. You are asking me what the world will be like in ten years, and I really have no clue what the job market will be like in ten years. I very strongly suspect that I won't be doing what I'm doing now. It could blow up in which case I'll be looking for whatever work I can (been there done that), or it could be spectacularly successful, in which case I'll get bored/cash out and the spend the rest of my life studying physics.

 

[icomeinpeace]

Thanks a lot for your prompt reply. It’s very kind of you to take time out of your busy schedule to post in this forum.

Actually, I wasn’t requesting advice for myself. I took computer science more than 20 years ago, and I’m quite happy with my choice.

I have a son who’s still in high school and likes math, computer science and business. Besides studying any of those three fields in particular, I thought the program I mentioned was a good combination of all three. In addition, I have another son already studying Physics at the undergraduate level. He’s the reason why I joined this forum – but that’s another story.

Like most Physics students, my son realizes he might not be able to continue onto graduate studies and/or to make a career out of it. So, I thought this thread could be of interest for both my sons. Just in case you’re wondering, my sons and I suggest stuff to read to one another, but we leave the decisions to each individual. Especially when it comes to career decisions :)

Also, the program I mentioned requires 5 subjects outside the Faculty of Mathematics. Students can not only take art, history, philosophy, or literature courses; but also, Latin, Religious Studies, “Sexuality, Marriage, and Family Studies”, and many others. I realize that 5 courses might not be broad enough for some people, but is certainly better than nothing. I’m not affiliated in any way with the University of Waterloo, but since I made it part of the discussion, I thought it was just fair to set the record straight.

Thanks again, and stay well,