Advice about Masters vs PhD ( twofish-quant)

[Cauchy1988]

Hi,
Many people have offered some great advice on these forums, and I specifically appreciate the well articulated advice from twofish-quant. I am about to enter graduate school with a focus on condensed matter theory but I am fairly certain that I don't want to do research as a career. Some options I've been pondering are becoming a quant or computer programming (although I'm just in the beginning of thinking about these things). I am wondering about whether to attain one of these jobs I should go ahead and complete the PhD, or would my time be better spent directly acquiring the skills these jobs require? By that I mean, should I not even bother with grad school and instead do things like take programming classes, seek an internship, etc? Or would something intermediate, like doing a Master's, be beneficial to my career or ultimately a waste of time?

 

[BenTheMan]

Do you WANT a phd in physics?

 

[Cauchy1988]

Do you WANT a phd in physics?

I don't want a PhD just for the sake of having a PhD but I'd be happy to go through the extra time and effort if it meant better opportunities/pay in becoming a quant later on. It seems that most jobs require at least a Masters, with PhD being preferred. Therefore I'd at least get the master's but whether to continue to get the PhD is still up in the air. Basically I am asking whether it is difficult to attain these jobs with just a master's. Or are employers are indifferent between a PhD and master's+a few years of experience of job experience in C++programming? Any advice would greatly be appreciated.

 

[BenTheMan]

twofish can offer an opinion from the inside. Let me be very clear that I am not an insider, but have a newly minted Ph.D. and am looking for a quant job.

Let me say this: if you WANT a PhD in physics, then you should get one. But if you don't WANT a PhD, then you should do something else. Seriously. It's like, would you want to be a Trappist monk just because you like beer? Sure, you get great beer, but do you really want to never use your sex organs again? (If I knew you were a dude, I would have said dick.)

If you WANT to be a quant, there are other ways to do it. You can get a PhD in finance, or economics. NYU has a master's program in Financial Math. My advice (which is an outsider, trying to get in, with a Ph.D. in physics) is you really weigh your options. If you're smart, a PhD won't make you smarter. If you don't believe me, go to LinkedIn and stalk people. Look for the kinds of backgrounds that people have who have job titles that appeal to you.

While it is true that a lot of physicists go to wall street, this is more for a lack of better options. I think that some people (like myself) go into finance because it is interesting. While I wouldn't trade my Ph.D. for anything, if I had it all to do over again, I'd do a Ph.D. in economics or finance, because I feel it is AS interesting as string theory. And a Ph.D. in economics gives you more options on Wall Street, especially if you can code in C++, C#, or Java.

So wait for twofish's advice. I think I know how he will answer your question (I THINK), but from someone who is going through the process now, you should be very careful about doing a Ph.D. in physics if you don't truly WANT a Ph.D. in physics.

That's just my two cents.

 

[BenTheMan]

Let me ask a related question: would you be willing to do a master's degree and then a Ph.D. in Finance? A friend of mine did exactly that.

 

[insom89]

I've read several other related posts and this keeps coming up; how is finance related to physics?

 

[Ben Espen]

Physics is not related to finance, however it turns out that you can model the price of financial instruments using techniques developed for physics, usually computational fluid dynamics. Physicists often have the required technical skills to create and maintain these models.

 

[physics8726]

Hi,
Many people have offered some great advice on these forums, and I specifically appreciate the well articulated advice from twofish-quant. I am about to enter graduate school with a focus on condensed matter theory but I am fairly certain that I don't want to do research as a career. Some options I've been pondering are becoming a quant or computer programming (although I'm just in the beginning of thinking about these things). I am wondering about whether to attain one of these jobs I should go ahead and complete the PhD, or would my time be better spent directly acquiring the skills these jobs require? By that I mean, should I not even bother with grad school and instead do things like take programming classes, seek an internship, etc? Or would something intermediate, like doing a Master's, be beneficial to my career or ultimately a waste of time?

Hi, I have a similar question. I want to apply a PhD program in finance in the future, and right now I have two choices: finish double major in physics and applied math or finish a master degree in physics. I can only finish one of them. Which one is more helpful for my plan in the future.

 

[twofish-quant]

I don't want a PhD just for the sake of having a PhD but I'd be happy to go through the extra time and effort if it meant better opportunities/pay in becoming a quant later on.

Don't do a Ph.D. if you don't want to do a Ph.D.

1) You can get equivalent experience elsewhere.
2) If you dislike doing Ph.D. work, then you'll hate doing it for the rest of your life.
3) Ph.D. take five to seven years, by the who knows what the market will be like.

If you want to do a Ph.D. for the sake of the Ph.D., then working in finance is a good way of getting paid well. However, if you don't have some inherent love for the topic, there are other ways of making money.

Or are employers are indifferent between a PhD and master's+a few years of experience of job experience in C++programming? Any advice would greatly be appreciated.

It's not so much indifference that there are enough jobs in both categories so that right now you can get in with either. Also you run a lot less risk with a masters since who knows what the economy is going to be like when you get the Ph.D.

 

[twofish-quant]

I've read several other related posts and this keeps coming up; how is finance related to physics?

Same sorts of equations and numbers. It turns out that the equation for modelling option prices is very similar to the equation that models the diffusion of neutrinos in supernova.

Also, the number of jobs in which you can use physics skills in finance is actually rather small. I'd have to say that 98% of jobs in finance have really nothing much to do with physics. But there aren't that many physics Ph.D. out there so the 2% of jobs that exist keep people employed.

Just to throw some numbers. There are about 800,000 people employed in the US in the securities industry. Each year, US schools produce about 1000-2000 new physics Ph.D.'s and about 120,000 MBA's.

 

[twofish-quant]

Hi, I have a similar question. I want to apply a PhD program in finance in the future, and right now I have two choices: finish double major in physics and applied math or finish a master degree in physics. I can only finish one of them. Which one is more helpful for my plan in the future.

They both will work. One problem with "which is better" questions is that often they both will work.

 

[twofish-quant]

I am about to enter graduate school with a focus on condensed matter theory but I am fairly certain that I don't want to do research as a career.

1) If you don't want to do research as a career, then you don't want to be a quant. What quants do is applied research, and it's really not that hugely different from graduate school except for the money, and even the money is a problematic.

One thing weird about human psychology is that people compare themselves to the people they are surrounded by, and if you go into finance, you are going to feel poor. You'll make 2x as much as if you did something else, but you will be interacting with people that make 4x or 10x.

2) If you have started something, then you need to finish it. If you are already a few months into a Ph.D. program, then you absolutely need to accomplish something (like getting your masters) before you leave.

I am wondering about whether to attain one of these jobs I should go ahead and complete the PhD, or would my time be better spent directly acquiring the skills these jobs require?

When a bank hires a physics Ph.D. it's because the Ph.D. has precisely the skill set that the bank wants. If they bank wants an MBA (and for 90% of the jobs in finance, it's better to hire an MBA), they would hire an MBA.

One important piece of advice for physics Ph.D.'s looking for jobs on Wall Street is to be a physics Ph.D. You aren't an MBA or Masters in Finance, and if you try to be one, you are going to be crushed by people who are.

By that I mean, should I not even bother with grad school and instead do things like take programming classes, seek an internship, etc? Or would something intermediate, like doing a Master's, be beneficial to my career or ultimately a waste of time?

If you do something computational, you will get more relevant computer experience in the physics Ph.D. than you will taking computer courses.

Also whether something is a waste of time or not depends on what you want to do with your life.

 

[DarrenM]

One important piece of advice for physics Ph.D.'s looking for jobs on Wall Street is to be a physics Ph.D. You aren't an MBA or Masters in Finance, and if you try to be one, you are going to be crushed by people who are.

I take your meaning here, but would you mind elaborating on this point a bit? I'm just curious.

 

[twofish-quant]

I take your meaning here, but would you mind elaborating on this point a bit? I'm just curious.

You'd naively think that if as a physics Ph.D. you want a job in a bank and you have to choice between researching magnetohydrodynamic models of black hole accretion disks or two factor interest rate models, that you should to the latter.

In fact, you end up more attractive to employers if you've done work on the first topic. It's because if they wanted an expert in two factor interest rate models, they wouldn't even look at you, since there are people that have ten times the knowledge on interest rate models than you. The reason you as a physics Ph.D. are interesting is because you've been researching MHD accretion disk models, and if you've done that, and read just enough about interest rate models to talk intelligently to someone that is the expert there, then this is what they are looking for.

 

[DarrenM]

Thank you. Most enlightening. Would you say that the same is true for a Math PhD?

If I'm not mistaken, twofish, you studied physics and eventually found yourself in financel; I know you can't necessarily speak for a Math PhD but I was wondering if you had encountered anyone post-academia that had made a similar transition.

 

[elkement]

Would you say that the same is true for a Math PhD?

I would say yes! I am a physicist working in IT security, basically implementing stuff based on cryptography. I meet a lot of physics and mathematics PhDs in this area.

Just as twofish-quant is often asked what finance has to do with physics I always get similar questions on the relation between physics and 'IT'. But for me this was a natural transition and I see lots of 'isomorphisms' between the fields (and even interfaces, consider quantum cryptography).

 

[twofish-quant]

Thank you. Most enlightening. Would you say that the same is true for a Math PhD?

Pretty much. The one caveat is that it helps to have broad knowledge. It helps in an interview if you've read enough to know what a "greek" is, and be able to program basic C++and matlab, but if you aren't an expert at this then you don't need to try to be an expert, because that's not why they are hiring you.

Also it works the other way. There are a lot of high paying jobs that don't involve being a physics geek. I'm not terribly interested in those.

If I'm not mistaken, twofish, you studied physics and eventually found yourself in financel;

The one thing that I worry about the OP is when he said that he is interested in finance because he/she doesn't want a career in research. It's terribly important to figure out why he/she doesn't want a career in research. I got into this field precisely because I wanted a career in research and this is the closest thing to what I want to do that I could find.

I know you can't necessarily speak for a Math PhD but I was wondering if you had encountered anyone post-academia that had made a similar transition.

Lots. The one difference is that in academia theory is good and applied is bad, whereas in industry theory is bad, and applied is good.

 

[DarrenM]

Lots. The one difference is that in academia theory is good and applied is bad, whereas in industry theory is bad, and applied is good.

Is that to say that you've encountered more Applied Math PhDs and fewer 'pure' Math PhDs? Or simply that the emphasis differs between academia and industry?

 

[twofish-quant]

Is that to say that you've encountered more Applied Math PhDs and fewer 'pure' Math PhDs? Or simply that the emphasis differs between academia and industry?

It's a marketing thing. In an academic situation, you often get people to bow down before you by talking about how abstract your research is. In industrial situations, you get people to bow down before you by talking about how applied your research is.

A lot depends on how you can sell yourself, and one problem that I've seen in Ph.D.'s is that they have been in an environment where they have been trained to think the "more abstract the better."

Also, I've seen people in the business with applied Ph.D.'s and people with pure Ph.D.'s. Having a Ph.D. in pure math won't hurt you, but having a personality that doesn't like "getting your fingers dirty" will cause problems.

 

[hadsed]

I would say yes! I am a physicist working in IT security, basically implementing stuff based on cryptography. I meet a lot of physics and mathematics PhDs in this area.

Just as twofish-quant is often asked what finance has to do with physics I always get similar questions on the relation between physics and 'IT'. But for me this was a natural transition and I see lots of 'isomorphisms' between the fields (and even interfaces, consider quantum cryptography).

I always thought that there had ot be some more interesting things in IT than just being a sysadmin. Can you elaborate on the type of work you do or see other physics Ph.D's getting to work on? I'm not a Ph.D student yet, but I want to know what the options are out there in the [probable] case that I decide academia just isn't worth the time. I've always been very interested in parts of computer science, quantum computing, anything that could be physics + CS. Like two-fish mentioned, I'd like to be able to do something like research, so being part of an R&D team at some company is perfectly okay with me, doesn't matter if it's aerospace, finance, IT/networking, or whatever.

 

[elkement]

I always thought that there had ot be some more interesting things in IT than just being a sysadmin. Can you elaborate on the type of work you do or see other physics Ph.D's getting to work on?

I am working in IT security - in the design, evaluation and implementation of systems based on public key cryptography. I feel that the number of colleagues with a physics or mathematics PhD is paticularly high in that field. The aspects that appeal to me in that area are
- mathematics, cryptopgraphy
- forensics, 'hacking', network sniffing, reverse engineering, trying to drill down to the details. This requires in my point of view not only 'IT knowhow' but some out-of-the-box thinking that is very 'physics like'. You can launch an attack against a cryptographic system by measurung the time dependence of electric power consumption by the processor, for example.
- working with chipcards. My PhD thesis was about superconducting films. I have some colleagues in IT security that worked on condensed matter physics, semiconductor physics or similar before.
- I am evaluating systems, e.g. try to find out why a protocol is secure or not and if adheres to a standard. I had also been in touch with teams working von quantum cryptography.

After some years of being heavily involved in all aspects of deployment (incl. troubleshooting) I stepped back a bit and became rather a reviewer and a consultant only.

Generelly I believe in the long run you should avoid IT jobs that are too much concerned with operations (like sysadmin, but the pressure and work-life-balance ist not much better if you are the IT manager).

If you want to work in R&D, the R&D department of a company that develops hardware-based cryptographic devices might be interesting for you, e.g. hardware security modules, smartcards and probably also quantum cryptographic devices. There are some startups that already sell commercial quantum crypto devices.

I see another upcoming and very interesting combination of physics and IT: classical building infrastructure and electrical power lines are more and more converging with classical IT infrastruture (incl. also consumer electronics - anything is converging, but there are still lots of hurdles in terms of standards). I think building these systems of the future will require interdisclinary know-how from IT and engineering / physics. As an IT security expert e.g. I am personally interesting in questions like 'how to hack / protect the smartgrid'.

 

[Goldbeetle]

If I were a physicist wanting to get into finance I would take a Ms in Economics - Finance.

 

[twofish-quant]

If I were a physicist wanting to get into finance I would take a Ms in Economics - Finance.

Bad idea.

The MS in Economics/Finance isn't a bad route for "some random person that wants to get into finance." It's a seriously bad route for a physicist that wants to do physics like things in finance for the money.

 

[hadsed]

I am working in IT security - in the design, evaluation and implementation of systems based on public key cryptography. I feel that the number of colleagues with a physics or mathematics PhD is paticularly high in that field. The aspects that appeal to me in that area are
- mathematics, cryptopgraphy
- forensics, 'hacking', network sniffing, reverse engineering, trying to drill down to the details. This requires in my point of view not only 'IT knowhow' but some out-of-the-box thinking that is very 'physics like'. You can launch an attack against a cryptographic system by measurung the time dependence of electric power consumption by the processor, for example.
- working with chipcards. My PhD thesis was about superconducting films. I have some colleagues in IT security that worked on condensed matter physics, semiconductor physics or similar before.
- I am evaluating systems, e.g. try to find out why a protocol is secure or not and if adheres to a standard. I had also been in touch with teams working von quantum cryptography.

After some years of being heavily involved in all aspects of deployment (incl. troubleshooting) I stepped back a bit and became rather a reviewer and a consultant only.

Generelly I believe in the long run you should avoid IT jobs that are too much concerned with operations (like sysadmin, but the pressure and work-life-balance ist not much better if you are the IT manager).

If you want to work in R&D, the R&D department of a company that develops hardware-based cryptographic devices might be interesting for you, e.g. hardware security modules, smartcards and probably also quantum cryptographic devices. There are some startups that already sell commercial quantum crypto devices.

I see another upcoming and very interesting combination of physics and IT: classical building infrastructure and electrical power lines are more and more converging with classical IT infrastruture (incl. also consumer electronics - anything is converging, but there are still lots of hurdles in terms of standards). I think building these systems of the future will require interdisclinary know-how from IT and engineering / physics. As an IT security expert e.g. I am personally interesting in questions like 'how to hack / protect the smartgrid'.

That sounds really interesting. My most recent feelings for what field I want to do my Ph.D in have been quantum computing or quantum information theory. I figured that this ought to be a really lucrative thing to do for the coming future, and it looks like there is some stuff I can still do if I can't work on R&D of quantum computers specifically. So in quantum information theory I assume that you learn the things needed for working on [quantum] cryptography and other sorts of things like you mentioned?

Sorry to hijack the thread, but there doesn't seem to be a whole lot of information about quantum computing or QIT in general, so I figure I should ask when I get the chance.

 

[elkement]

So in quantum information theory I assume that you learn the things needed for working on [quantum] cryptography and other sorts of things like you mentioned?

Most 'quantum cryptographers' I know did not have a specific training in quantum information / quantum cryptography. They rather gradually mved into that field. They came from related classical disciplines - as diverse as laser physics (required for building the devices), loosely related areas in theoretical physics, mathematics and classical cryptography (IT, not physics), superconducting devices...

So the field is truly interdisciplinary and I think it is most important to find a research group working on quantum technologies and start working in field early. I would do a survey on companies working in that area and check out what backgroud the people there have: E.g. I recently found that company on the web, quite cool I think: http://www.dwavesys.com/en/technology.html
They also have a blog (http://dwave.wordpress.com/) that could give you a realistic picture of what you do a 'quantum engineer'.

 

[hadsed]

Wow. It's really nice to see start-up companies in quantum computing already.

One more thing I wanted to ask about. I'm highly inclined towards more theoretical and computational aspects of problems, not so much experimental. Is this an advantage or disadvantage do you think? Although I guess I'd imagine that depends on what field I'm looking at, but you mentioned your doctorate was in superconducting films, and others have had experience in what sounds more like experimental disciplines than theoretical. What do you think?

 

[James A. Putnam]

Cauchy1988,

If you are still around here: A phd is about learning significantly more than others know. It is about continuing your education. If you have the desire and ability, then, get your phd. If you wish to take a job where you think that having a phd would disqualify you then do not say that you have a phd. You will not be the first person to say either that or that you do not have a masters degree. Knowing as much as you can will separate you from others. In the corporate world, when project managers are wanted, the corporations want applicants with phds. Earning a phd does not necessarily mean doing research. It may mean leading others while solving problems even the researchers have not solved. The corporate world embraces competent people. Having a phd does not guarantee competence in the corporate world. Get whatever education you can afford. Then determine what you are competent at. The academic world can be but is often not the best place to determine that. Sometimes the corporate world can be more freeing than the academic world. The corporate world has to produce results while the academic world sometimes is complacent with the status quo. What I mean is that change can become forced more rapidly in the corporate world out of necessity. What is it that you would really like to be doing? If you are presently continuing your education, then this question can wait. You are most likely already being driven in that general direction.

James

 

[elkement]

Wow. It's really nice to see start-up companies in quantum computing already.
One more thing I wanted to ask about. I'm highly inclined towards more theoretical and computational aspects of problems, not so much experimental. Is this an advantage or disadvantage do you think? Although I guess I'd imagine that depends on what field I'm looking at, but you mentioned your doctorate was in superconducting films, and others have had experience in what sounds more like experimental disciplines than theoretical. What do you think?

Difficult so say - I can only provide anecdotal evidence. But generally I believe it is not so much the classification of your thesis (experimental / theoretical) or which lectures you have attended. If you want to work in industry, but specifically if you want to work for a startup, it is more important that you can prove that your attitude is 'experimental' / hands-on. You should not be afraid of 'getting your fingers dirty' and tackle whatever problem is here to be solved. This does not apply to lab work only. Working in a startup or small company in generall means that you might be required to work in different roles, e.g. you would not only develop the fancy new device but probably also do project controlling or contribute to the marketing plan.

Along these lines I will also try to respond to the initial question in the thread (Sorry Cauchy1988, I did not intend to hijack the thread, but just wanted to respond to other questions coming up)

Hi,
Many people have offered some great advice on these forums, and I specifically appreciate the well articulated advice from twofish-quant. I am about to enter graduate school with a focus on condensed matter theory but I am fairly certain that I don't want to do research as a career. Some options I've been pondering are becoming a quant or computer programming (although I'm just in the beginning of thinking about these things). I am wondering about whether to attain one of these jobs I should go ahead and complete the PhD, or would my time be better spent directly acquiring the skills these jobs require? By that I mean, should I not even bother with grad school and instead do things like take programming classes, seek an internship, etc? Or would something intermediate, like doing a Master's, be beneficial to my career or ultimately a waste of time?

I can share experiences from IT industry, thus related to programming not to finance. I believe that for somebody considerng to hire you as a programmer it does not matter so much if you have a PhD or 'only' a master. It is the skills you have acquired that count. Personally I would try to combine working on a PhD AND obtaining software development skills - consider your PhD your first real job. I had a colleague at the university who was happily working on his PhD in experimental physics but who said from day 1 that he would want to work as a software developer later.

There is a slight chance that potential employers might consider you 'too acedemic' though. It depends on the corporate culture, but you might be interviewed by somebody who started his career as the classical hacker who learned anything on his own and who considers academic or professional titles just a show-off.
If you want to prove to those people that you will be useful you need to convince them that - as I said before to hadsed - you get your hands dirty and find pragmatic solutions ('although' you have a PhD).

Responding to both of you (hadsed, Cauchy1988):
I was probably really lucky that my thesis advisor forced me to do not only scientific / bench work, but also project management. So I could explain to potential employers later that this was really my first job and not some extended stay in what they consider the academic ivory tower.

 

[James A. Putnam]

Cauchy1988,

My second job, second because the first changed its corporate status, was in a startup company. We designed microcircuits in the very early seventies. There were different technologies having different advantages. The circuit designers had to make the difference. It was good and it was bad. The good was that innovation really mattered. I was innovative.

The bad was that computers were so slow, that real scientific use was only possible at nightimes. Also, innovation does not know clock time. Long days and nights were often necessary. The point is that, even with super computers today, innovation can be slow in coming. Innovators still have to wait to get cumputer time.

My final point is that in the end it is innovation that matters for science. It is ironic that innovation mayn sometimes be blocked even by science. If you are still around. Stay your course. Be innovative and that includes finding platforms for innovation. Sometimes innovators must provide their own platforms for innovation.

James