Programs Do you need a PhD to become a quant?

[hatelove]

To be a Derivatives Quant, Stats quant, etc.

 

[Totalderiv]

Most quant jobs require a PhD, so I would say yes.

 

[twofish-quant]

I don't know what a quant is. (Seriously, I don't.)

90% of the people in finance with "quant" in their title are high end computer programmers. About 60% of the people I work with have Ph.d.'s and the other 40% have masters in some technical field with a lot of work experience before they got hired in finance.

Also if your main interest is finance or to make money, do *NOT* get a Ph.D. in physics. It's silly to get a Ph.D. in physics if your main interest is either to work in finance or make money, for many reasons, not the least of which is that no one has a clue what the world financial situation will look like in five years.

Now if your main interest is physics and finance is just a way of paying the bills, that's a different issue.

 

[hatelove]

90% of the people in finance with "quant" in their title are high end computer programmers. About 60% of the people I work with have Ph.d.'s and the other 40% have masters in some technical field with a lot of work experience before they got hired in finance.

Ah, I had no idea. I thought they mainly worked with finance, and the programming was minimal.

I read this blog entry: http://getonthedesk.blogspot.com/2008/01/day-in-life-of-quant.html
And I don't know how accurate it is, but it seemed like the programming was consisted of short coding periods (I can write up small programs [which might even be just considered scripts] too, though I wasn't aware how heavily programming was integrated in these types of careers).

Also if your main interest is finance or to make money, do *NOT* get a Ph.D. in physics. It's silly to get a Ph.D. in physics if your main interest is either to work in finance or make money, for many reasons, not the least of which is that no one has a clue what the world financial situation will look like in five years.

So a Physics PhD would not be versatile at all in case of financial emergency (if, God forbid, a terrible calamity occurred with the economy in which a particular degree might be rendered null or something)?

Can you suggest more suitable degrees aside from, say, a Physics undergrad + Physics PhD?

 

[hatelove]

I'm just looking over some general job descriptions/qualifications as examples here:

http://careers.bankofamerica.com/JobDetails.aspx?SearchPage=ASP&CountryId=1&JobId=1100066493&stateid=-1&cityid=-1&travel=-1&jfamily=&lang=&fullpart=&shift=&datepost=0&keywords=quant&jobareas=-1|-1&feedname=BOAFEEDUSA&BOAFeedName=&jobfamilyid=0&internationalcity=&LocationID=0

http://careers.bankofamerica.com/JobDetails.aspx?SearchPage=ASP&CountryId=1&JobId=1100057558&stateid=-1&cityid=-1&travel=-1&jfamily=&lang=&fullpart=&shift=&datepost=0&keywords=quant&jobareas=-1|-1&feedname=BOAFEEDUSA&BOAFeedName=&jobfamilyid=0&internationalcity=&LocationID=0

http://careers.bankofamerica.com/JobDetails.aspx?SearchPage=ASP&CountryId=1&JobId=1100067738&stateid=-1&cityid=-1&travel=-1&jfamily=&lang=&fullpart=&shift=&datepost=0&keywords=quant&jobareas=-1|-1&feedname=BOAFEEDUSA&BOAFeedName=&jobfamilyid=0&internationalcity=&LocationID=0

They require PhDs in "Quantitative" fields, which I'd presume includes Physics as well (Physics is just the first thing that comes to mind when "quantitative" is mentioned; I think they do list Physics as an example)

 

[hatelove]

I feel so discouraged when I view the CVs of students like these:

http://www.princeton.edu/bcf/graduate/recruiting/graduating/

Will I even have a shot in this field with people like these running the show? Their resumes seem padded, but they also look very related and impressive, not to mention all the other hot shots that will be graduating from other top schools...

 

[twofish-quant]

Ah, I had no idea. I thought they mainly worked with finance, and the programming was minimal.

Most jobs in finance don't require much programming, but quantitative jobs make up a tiny, tiny fraction of jobs in finance.

I read this blog entry: http://getonthedesk.blogspot.com/2008/01/day-in-life-of-quant.html
And I don't know how accurate it is, but it seemed like the programming was consisted of short coding periods (I can write up small programs [which might even be just considered scripts] too, though I wasn't aware how heavily programming was integrated in these types of careers).

It's reasonably accurate for someone doing front desk support, but that's only one type of job, and it's not a job that's in much demand now. Most of the hiring that I've seen has been in risk management and infrastructure programming.

Also, the fact that someone spends only a two hours coding means that you have to be an extremely, extremely sharp programmer. In a production support issue, you are expected to go through a system with several hundred thousand lines of c++ code, diagnosis the problem within ten minutes, and then come up with a fix within an hour.

Those situations are a lot like being an emergency room doctor. If you look at the schedule of an ER doctor, they might spend only an hour trying to deal with someone with a heart attack but...

So a Physics PhD would not be versatile at all in case of financial emergency (if, God forbid, a terrible calamity occurred with the economy in which a particular degree might be rendered null or something)?

On the other hand, if the world economy collapses, than no degree is going to help you. Then again, being smart and curious can help you in extreme events.

Can you suggest more suitable degrees aside from, say, a Physics undergrad + Physics PhD?

This really is a personal decision, and any advice that I give is inherently self-defeating. If I tell everyone to get a degree in X, then soon there is going to be a glut in X. If you love physics, and are willing to go through a fair amount of pain to learn physics, then a physics Ph.D. is decent. If you don't care about physics, then there are *thousands* of better options.

 

[twofish-quant]

Their resumes seem padded, but they also look very related and impressive, not to mention all the other hot shots that will be graduating from other top schools...

For the type of work that I do, those resumes are very unimpressive, and there are only two or three of them that my group would even consider calling in for a phone screen. There is no evidence of any sort of deep mathematical or programming ability. The type of resume my group would be more interested in is someone who has worked for five years at Microsoft or Google or someone with deep applied mathematics, statistics, or engineering experience.

Among the people that I work with, you have a few people with masters of finance engineering or financial mathematics, but no one with a straight masters of finance, and we are more likely to hire a mechanical engineer than someone with a finance masters, because people that go the masters of finance route are generally not too interested in the "geek work" that my group does.

Now there are other groups in which programming and mathematical skills are not so important, and if one of those resumes came across my desk, I'd likely forward the resumes to them.

Also for the work that I do, "top schools" don't matter. Physics graduate schools are not strongly tiered, and the fact that finance is a second choice for physicists means that the resumes that come across tend not to come from a big name school.

 

[RugbyEng]

There are a ton of MFE programs out now. Maybe two-fish can comment on weight of a MFE from NYU or Columbia vs NCstate or Kent. In the early days a lot of PhD's would be hired because the field was blossoming and academic positions for physicist were becoming more rare post wartime. I believe the entry level quant positions are programming dominant, mainly in c++ and matlab, if I'm not mistaken. I also feel like the main goal of the entry level quant is to eventually obtain a position as a trader (where the real money lies). From what I've read most people in finance burn out within 5 years, so you need to decide if it is something you are really interested in.

I think I read that a lot of people who start as "quants" never really make it "big," and are really just coders.

 

[DeadOriginal]

There are a ton of MFE programs out now. Maybe two-fish can comment on weight of a MFE from NYU or Columbia vs NCstate or Kent. In the early days a lot of PhD's would be hired because the field was blossoming and academic positions for physicist were becoming more rare post wartime. I believe the entry level quant positions are programming dominant, mainly in c++ and matlab, if I'm not mistaken. I also feel like the main goal of the entry level quant is to eventually obtain a position as a trader (where the real money lies). From what I've read most people in finance burn out within 5 years, so you need to decide if it is something you are really interested in.

I think I read that a lot of people who start as "quants" never really make it "big," and are really just coders.

From what I've heard most quants with the sole goal of becoming a trader burn out within 5 years. Most people who shoot for the money don't make it too far.

Source: www.quantnet.com

 

[hatelove]

Most of the hiring that I've seen has been in risk management and infrastructure programming.

So does this naturally bring the focus back to something like a computer engineering degree, as opposed to something like a degree in physics? Which would actually be more preferred in your line of work?

Speaking of which, I'm not exactly sure what you do besides programming. Do you have a link to one of your posts which describe your job in detail?

On the other hand, if the world economy collapses, than no degree is going to help you. Then again, being smart and curious can help you in extreme events.

This is actually what I'd like to concentrate on more. I'm by no means the most natural of abilities when it comes to problem solving, but I'd like to learn and improve my breadth of knowledge by expanding my vision to see outside of the box. Again, speaking of versatility, I thought maybe physics would help me achieve this as well as being able to land me a decent job where I can put my learned theories into practice.

Maybe going into the field of mathematics might perform similarly? Though I'm not sure how viable that would be in comparison to physics or engineering...

 

[RugbyEng]

Personally, if I was going to try and be a quant, I would pick my favorite engineering and pursue a BS (just not civil), maybe minor in math or cmsc at the same time. Work your tail off and get over a 3.5, graduate, get over a 750 on the math portion of the GRE, and hope to get into one of those Ivy league MFE programs.

Here are Cal Berkley's prereq's:

Successful candidates for the MFE Program will have a strong background in
Computer programming
High-level math and statistics
Finance studies
Langauge skills
Statistical and econometric applications (Sas, Gauss, RATS, S-Plus, or Garch)
Mathematical tools (MatLab, Mathematica, or MathCad)
Before the program you are applying to is scheduled to begin, you should:
Have taken—or have a plan in place to take—the prerequisite courses listed below for a grade of "B" or higher
Plan to take all of the pre-program courses in addition to the prerequisites to reinforce your understanding of the basic concepts
Please note that you do not necessarily need to complete all of the coursework prior to submitting your application, but you do need to have a clear plan in place to complete the coursework between the time of application and the time the program begins.
For students who have not taken a math course in more than 5 years, we do recommend some type of refresher course in order to excel in the program.

Prerequisite Course List

Computer Programming Experience
Requirement:
Prior experience in computer programming (C, C++) and familiarity with computers as a computational and management tool.
Area:
C, C++ Programming
Suggestion:
1 course OR equivalent work experience
[Back to top]
Quantitative Background
Requirement:
A strong quantitative background including multivariate calculus, linear algebra, differential equations, numerical analysis and advanced statistics and probability.
Calculus
Linear Algebra
Differential Equations
Statistics
Numerical Analysis

Training in Finance

They want like 2 classes in finance or economics.

With the engineering degree all of the maths would be done except analysis, you could take some econ and finance classes as electives, and c++/matlab you could probably semi-learn in your engineering classes and your own time.

Now, the benefit of the engineering BS, if the financial world tanks again, you can get an engineering job.

http://www.quantnet.com/ this website has a lot of information, and if you're really interested you can pick up Emanuel Derman's "My Life as a Quant."

 

[gravenewworld]

Nah just a borderline criminal mindset

 

[hatelove]

Personally, if I was going to try and be a quant, I would pick my favorite engineering and pursue a BS (just not civil), maybe minor in math or cmsc at the same time. Work your tail off and get over a 3.5, graduate, get over a 750 on the math portion of the GRE, and hope to get into one of those Ivy league MFE programs.

Now, the benefit of the engineering BS, if the financial world tanks again, you can get an engineering job.

This is actually what I was most likely considering...I wanted to take MSE @ NC State and then work on my PhD, but it seems getting an MFE first would be more sound going into the finance sector. If I can finish an MFE within a year, maybe I can still continue my PhD education afterwards?

http://www.quantnet.com/ this website has a lot of information, and if you're really interested you can pick up Emanuel Derman's "My Life as a Quant."

Thanks, I've checked out that site and it would seem it's pretty helpful. Also makes me feel like the market is pretty saturated, but I'm going to go for it anyway.

 

[twofish-quant]

There are a ton of MFE programs out now. Maybe two-fish can comment on weight of a MFE from NYU or Columbia vs NCstate or Kent.

Personally, I'd avoid MFE programs and go for either a masters in computer science/statistics/MBA/finance or anything else. The problem with MFE programs is that you are dead if quantitative finance falls apart, whereas if you have a masters in something else, you have options other than Wall Street.

In the early days a lot of PhD's would be hired because the field was blossoming and academic positions for physicist were becoming more rare post wartime. I believe the entry level quant positions are programming dominant, mainly in c++ and matlab

Which works great for physics Ph.D.'s with computational skills.

if I'm not mistaken. I also feel like the main goal of the entry level quant is to eventually obtain a position as a trader (where the real money lies).

This is wrong. There is the occasionally quant that moves into trading, but it's rather uncommon since the skill sets are different. Also, there is less money in trading that you may think. It's true that head traders make a ton of money (if the market works) but you aren't going to be a head trader immediately, and you need to spend several years as a trading assistant, which makes a decent but not spectacular, amount of money.

From what I've read most people in finance burn out within 5 years, so you need to decide if it is something you are really interested in.

This tends not to happen to physics Ph.D.'s. Part of the reason that banks hire physics Ph.D.'s is that if you haven't burned out getting the Ph.D., you are unlikely to do so in a banking environment.

The stress is high, but no worse than graduate school.

I think I read that a lot of people who start as "quants" never really make it "big," and are really just coders.

95% of quants are glorified coders, but good coders make a lot of money.

 

[twofish-quant]

So does this naturally bring the focus back to something like a computer engineering degree, as opposed to something like a degree in physics? Which would actually be more preferred in your line of work?

No major preference. It's in fact a good thing to have a mix of people. Physicists learn a different set of techniques than mathematicians, statisticians, econometrics, and/or CS people, so within the scope of technical people, it's desirable to keep a diverse mix.

Maybe going into the field of mathematics might perform similarly? Though I'm not sure how viable that would be in comparison to physics or engineering...

Applied mathematics is a good way into the field. One thing to remember is that for the jobs that are of interest to people in this board, you aren't looking at a "job in finance" you are looking at a "job in applied mathematics or physics that happens to be in a financial institution."

 

[twofish-quant]

Personally, if I was going to try and be a quant, I would pick my favorite engineering and pursue a BS (just not civil), maybe minor in math or cmsc at the same time. Work your tail off and get over a 3.5, graduate, get over a 750 on the math portion of the GRE, and hope to get into one of those Ivy league MFE programs.

I wouldn't. MFE programs aren't particularly a good way of getting into "geek jobs" and they are extremely risky if the economy changes. Also *big name* MFE's aren't necessary worth their tuition. The one thing that you get from an MFE is career placement and there are "no name" MFE's with wonderful placement and "big name" MFE's with horrible placement.

Here are Cal Berkley's prereq's

Which is a bit silly. If you have all of this background then what the heck are you getting out of the MFE program? One thing that you have to be careful is the "placebo degree" which is a degree in which you pay massive amounts of money to get a job that you probably would have gotten anyway.

http://www.quantnet.com/ this website has a lot of information, and if you're really interested you can pick up Emanuel Derman's "My Life as a Quant."

You have to be careful about some of the information. Derman's book is a great book provided that you read it as *history* and realize that a lot of the information in that book is quite out of date.

 

[twofish-quant]

This is actually what I was most likely considering...I wanted to take MSE @ NC State and then work on my PhD, but it seems getting an MFE first would be more sound going into the finance sector. If I can finish an MFE within a year, maybe I can still continue my PhD education afterwards?

If you are going to get a Ph.D. get a Ph.D. If you get an MFE first it's going to be totally useless by the time you finish your Ph.D. The big thing that you get from the MFE is career placement and internships, and if you aren't intending to get hired very quickly, those are useless.

Thanks, I've checked out that site and it would seem it's pretty helpful. Also makes me feel like the market is pretty saturated, but I'm going to go for it anyway.

I have a generally bad impression about that site. The information is not particularly useful, and it seems more like ads for MFE programs.

One problem with MFE is that things change too quickly. If you pick up a book on business administration or applied math that was written in 2005, then most of it would still be useful today, but if you pick up any book on quantitative finance that was written in 2005 then most of it is either wrong or useless.

 

[hatelove]

Ah, I was actually planning a trip to the library on Tuesday to borrow some books off that list on the quantnet website. Thanks for the precautionary heads up.

Are there any specific texts that you'd personally suggest?

 

[hatelove]

I was looking for something cross between applied mathematics and statistics, and I've recently discovered something called "actuarial science..." This might be the best option I've found so far because I wanted to go into something like applied statistics (which I don't think exists) and actuarial science seems pretty close.

 

[twofish-quant]

Are there any specific texts that you'd personally suggest?

Start with Hull's Options, Futures, and other Derivatives and Paul Wilmott's book.

80% of both books are wrong or irrelevant, but it's useful to know the terminology so that when someone explains to you want a three factor interest rate bond model isn't useful or why a local volatility model won't work, you know what a three factor interest rate or local volatility model are.

The really cool problems don't have textbooks.

 

[chiro]

I was looking for something cross between applied mathematics and statistics, and I've recently discovered something called "actuarial science..." This might be the best option I've found so far because I wanted to go into something like applied statistics (which I don't think exists) and actuarial science seems pretty close.

Statistics is one of the most applied fields out there. It's almost an oxymoron to have statistics not be an applied fields (and yes I am aware of theoretical results in statistics and associated research).

If you want to do any kind of "applied" statistics, just major in it and do some post-graduate work in it along with as many projects as you can. Actuarial work is statistics applied to risk management, but there are other applications of statistics which include to the sciences (including the area of health) as well as other things like business and government.

 

[hatelove]

Statistics is one of the most applied fields out there. It's almost an oxymoron to have statistics not be an applied fields (and yes I am aware of theoretical results in statistics and associated research).

If you want to do any kind of "applied" statistics, just major in it and do some post-graduate work in it along with as many projects as you can. Actuarial work is statistics applied to risk management, but there are other applications of statistics which include to the sciences (including the area of health) as well as other things like business and government.

I was actually warned by several other people that any degree (especially a PhD) which is too "general" might not be such a good idea, and that I should specialize in a certain field. Does this apply to statistics as well, or will a general statistics degree be fine on the market?

 

[npc]

I have always been facinated by the number of people who seem to want do physics to become 'quants'. Everyone should realize that 'quant' work is not physics and never will be, so don't do physics just because you want to become a quant. The attraction is the money, but this is not going to last much longer.

Future job prospects are also not great, certainly, across europe now most organisations are deleveraging, getting rid of the complicated stuff and going more 'vanilla', the result of this will be less need for high end quant people.


Mind you, the quant who comes up with a derivative that avoids the proposed eurpoean financial transaction levy will have a secure job...

 

[kmgilroy89]

There are different kinds of quants. Quantitative research usually wants people with a PhD in math or physics, because of:
1. High level math involved (stochastic processes, numerical analysis, PDE)
2. Research experience

 

[Kevlarji]

I say no, I've been to numerous bulge bracket bank campus visits and have asked them this question. "Do I require a PhD to get into Quant?" the response majority of responses I've had is no you can get into an analyst position but if you make it clear you are very interested in quant finance you can leverage yourself into a quant role without having a PhD and getting in the general way.

 

[twofish-quant]

I was actually warned by several other people that any degree (especially a PhD) which is too "general" might not be such a good idea, and that I should specialize in a certain field. Does this apply to statistics as well, or will a general statistics degree be fine on the market?

One problem with advice is that you'll end up with different people telling you to do completely different things, and so you'll have to work things out for yourself.

For me:

1) I got my Ph.d. for the sake of getting the Ph.d.

2) Now once I got my Ph.D., I found is was extremely useful. For me "general" is good, because my career has involved moving from field to field looking for where ever the grass is greenest. The problem with specializing in a specific field is that you are totally hosed if everyone else decides to do the same thing.

My guess is that ten years from now, I won't be doing finance. Even if everything works great, I'll get bored and want to do something else.

 

[twofish-quant]

I have always been facinated by the number of people who seem to want do physics to become 'quants'. Everyone should realize that 'quant' work is not physics and never will be, so don't do physics just because you want to become a quant.

There are about 20 different jobs that qualify as "quant." It so happens that the work that I do *is* pretty close to physics. In graduate school, I spent 12 hours/day working on numerical code. Today, I do pretty much the same thing and crunch pretty much the same equations, and do the same sort of modelling.

The attraction is the money, but this is not going to last much longer.

That's not the attraction for me. It helps, but you can make decent amounts of money working elsewhere.

Future job prospects are also not great, certainly, across europe now most organisations are deleveraging, getting rid of the complicated stuff and going more 'vanilla', the result of this will be less need for high end quant people.

1) Europe is shrinking but other parts of the world are growing

2) Exotics derivatives are dead, but there are a dozen other interesting research areas (risk management, and algo trading). The cool thing about the field is that you can get a standard textbook on exotic derivative pricing. There is no standard textbook technique for measure bank liquidity.

Mind you, the quant who comes up with a derivative that avoids the proposed eurpoean financial transaction levy will have a secure job...

I do know people that do that (i.e. given regulation X, figure out strategies that minimize the financial impact of regulation X).

 

[twofish-quant]

I say no, I've been to numerous bulge bracket bank campus visits and have asked them this question. "Do I require a PhD to get into Quant?" the response majority of responses I've had is no you can get into an analyst position but if you make it clear you are very interested in quant finance you can leverage yourself into a quant role without having a PhD and getting in the general way.

1) You do need a masters degree in something. I don't know of anyone that has gotten hired with a bachelors.

2) Because the term "quant" is cool, every job has "quant" associated with it. Most Ph.D. level "quants" are just glorified programmers, and there is a ton of masters-level "quants" that are just data entry people.

3) Don't be blinded by the money. It's good, but realistically, it's not that good. The salaries get a lot less impressive once you figure out that you are living in wildly expensive cities. If you live in NYC, you are dealing with a 45% marginal tax rate which means that almost half your bonus goes to the government.

 

[hatelove]

2) Now once I got my Ph.D., I found is was extremely useful. For me "general" is good, because my career has involved moving from field to field looking for where ever the grass is greenest. The problem with specializing in a specific field is that you are totally hosed if everyone else decides to do the same thing.

My guess is that ten years from now, I won't be doing finance. Even if everything works great, I'll get bored and want to do something else.

Well, if you're living proof that your degree isn't unemployable, then I don't feel so worried after all.

I'm writing up a list for reading and I've included Hull's Options, Futures, and other Derivatives and the books of Paul Wilmott that I found on his Wikipedia page.

 

[Poopsilon]

Twofish-quant, you give a lot of contrarian advice in this thread, which isn't necessarily a bad thing, but is not exactly actionable.

Thus maybe you could talk about what someone with a masters in math and a strong foundation in statistics, analysis, probability, and programming can do to further develop his or her skill set in order to make themselves competitive for those math heavy positions, (note I won't use the word quant since you say you don't know what that word means).

Will it be just to continue to master increasingly sophisticated areas of math which have been historically associated with quantitative finance?

I don't think anyone wants to waste time learning material which will be useless in a few years when they apply. I understand that when you have the job then part of it will be continuously learning new material as the industry evolves. But for those who don't have the job yet, how can we develop ourselves to stand out without spending 60k on an MFE or 5 years getting a phd?

 

[chiro]

I was actually warned by several other people that any degree (especially a PhD) which is too "general" might not be such a good idea, and that I should specialize in a certain field. Does this apply to statistics as well, or will a general statistics degree be fine on the market?

It's like anything else: you need to get real applicable experience. This is why you do projects, get internships and the like.

My professor who taught experimental design was telling the story of how he had all these neat mathematics like measure theory in his honors degree but found out he didn't real know that much when he was working out in the work-force.

The mathematics was not a substitute for working on real projects and building up an intuition in that regard. I dare say this is more of a template than an anomaly.

The best thing you can do if you want to pick something that is applicable is to pick subjects where you work on projects of some sort. Working on projects with other people is better and if you ever get the chance to work on interdisciplinary projects (i.e. with people that have other skillsets and knowledge basis), then that is even better.

A lot of classes in certain disciplines will force you to do projects, work with other people and write up reports. This is the kind of training for a wide range of technical jobs that go outside of academia (as well as academia if need be).

 

[twofish-quant]

Twofish-quant, you give a lot of contrarian advice about becoming a quant, which isn't necessarily a bad thing, but is not exactly actionable.

I just tell you what I'm seeing.

Thus maybe you could talk about what someone with a masters in math and a strong foundation in statistics, analysis, probability, and programming can do to further develop his or her skill set in order to make themselves competitive for those math heavy positions, (note I won't use the word quant since you say you don't know what that word means).

I think that's the wrong question. One thing that you have to realize is that if things become too competitive that means that there are too many applicants and too few jobs, and if you find yourself in that situation, you need to seriously ask if you want to do something else.

Also, I really don't know which skills are going to be in demand in three years. The way I've dealt with this is to be flexible so that I can quickly learn whatever it is that I need to learn. The other thing to do is to reverse the problem, I have skills X, Y, and Z, how do I market X, Y, and Z?

Will it be just to continue to master increasingly sophisticated areas of math which have been historically associated with quantitative finance?

The math that was useful three years ago turns out to be useless today. People right now aren't particularly interested in stochastic differential equations or copula models, for example.

This is great if you like learning new stuff all of time. Sort of stinks if you are in the "get the degree, get the paycheck" mode of thinking.

I don't think anyone wants to waste time learning material which will be useless in a few years when they apply.

It's not a waste. I like learning math and physics for the sake of learning math and physics. Also, doing any sort of hard math helps you quickly learn some other hard math. For example, suppose you are a hot shot at algebraic topology or quantum field theory. It turns out that those are not obviously useful for finance. But... If you have someone that can to algebraic topology or QFT and better yet *likes* to do topology or QFT, then you can give them Hull, and then can learn what they need to know in a week or two.

Something that happens quite often is that people ask me to work on a computer program in a language that I've never seen concerning math that I've never studied on a finance model that I've never seen before, and I have a week to fix the problem.

I think that's cool.

If you are in the "I just want to learn the minimal stuff to get paid" then you are going to hate doing my job.

I understand that when you have the job then part of it will be continuously learning new material as the industry evolves. But for those who don't have the job yet, how can we develop ourselves to stand out without spending 60k on an MFE or 5 years getting a phd?

One thing that I've found is that if you have to "stand out" to get a job, then you might consider working for another field.

One thing about me is that I never intended to go into finance. My main goal in life is physics and math. It turns out that by random chance, I ended up in finance, but if something else was closer to what I wanted, I would have ended up there.

 

[RugbyEng]

What's wrong with a BS in Engineering and an MFE? The MFE will get your foot in the door and the BS is marketable if the economy turns sour. Just make sure you know your C++ and Matlab. Or you could go BS in Stat or Applied Math.

 

[Poopsilon]

I'm sorry but this advice is still next to useless, and is essentially a rehashing of what you have always said in this post and in others. I'm not looking for a life coach, and I don't think anyone else here is either.

I personally am trying to keep a number of irons in the fire, finance being one of them, and if it turns out that the market is over saturated, well then I will go somewhere else.

And before you warn me about the dangers of keeping too many irons in the fire and spreading myself too thin, assuming someone has decided they want to have a job like yours, what are the right questions to be asking? And while you're at it maybe you could answer a few of them as well.

 

[chiro]

What's wrong with a BS in Engineering and an MFE? The MFE will get your foot in the door and the BS is marketable if the economy turns sour. Just make sure you know your C++ and Matlab. Or you could go BS in Stat or Applied Math.

The thing is that with some of these kinds of jobs, your job is to 'figure out' the rules instead of applying them.

As twofish-quant has pointed out in other threads, the rules in finance change every couple of months at times while the laws of physics don't: which is why for some roles PhD candidates are preferred since what they do is akin to what needs to be done in the job.

Also for programming, it takes a long time to be able to have both the breadth and specificity required in doing things effeciently. I used to be a programmer so I have had this experience directly and if you haven't had a lot of experience on real projects then people will see through you when it comes interview time should you get that far.

I feel sorry for twofish having to say the same thing over and over: maybe he should make a stickied post for this kind of thing so he doesn't have to spend time saying the same thing over and over again

 

[RugbyEng]

I'm sorry but this advice is still next to useless, and is essentially a rehashing of what you have always said in this post and in others. I'm not looking for a life coach, and I don't think anyone else here is either.

I personally am trying to keep a number of irons in the fire, finance being one of them, and if it turns out that the market is over saturated, well then I will go somewhere else.

And before you warn me about the dangers of keeping too many irons in the fire and spreading myself too thin, assuming someone has decided they want to have a job like yours, what are the right questions to be asking? And while you're at it maybe you could answer a few of them as well.

I merely provided an example of an alternative route, no need for the negativity.

An example job posting, it tells you exactly what they are looking for. We are all pretty smart people we can figure out what degrees and training will give us these qualifications.

Job Requirements: (Knowledge/ Experience):
* 3-5 years of mid/high frequency quantitative FI trading experience
* Science focussed degree level education
* Significant programming experience
* Quantitative and analytic skills
* Strong interpersonal skills
Required Skills/ Competencies/ Attributes:
See Above
Education:
BA/BSc

Obviously, there are multiple ways to achieve these qualifications and PhD is not required (to answer the OP's question). A masters most likely is, as two-fish pointed out earlier in the thread. Also, the only degree that is really super-focused on quantitative finance is the MFE, so if you go another route you can keep your "irons in the fire." At the same time the MFE has the benefit of internships and more exposure. Thus, each route has it's pros and cons and its up to the individual to decide which one is best for him/her.

Two-fish, would you say with the multitude of MFE programs out there and the rising recognition of the "quant career field," (saturation) that the hardest part of becoming a quant is getting your foot in the door?

 

[twofish-quant]

I'm sorry but this advice is still next to useless, and is essentially a rehashing of what you have always said in this post and in others. I'm not looking for a life coach, and I don't think anyone else here is either.

Career is a very important part of your life. You really can't figure out what you want to do with your career if you don't know what you want to do with your life.

If you want a cookbook recipe for making money, I can't give you one, because I'm reasonably sure that one doesn't exist.

And before you warn me about the dangers of keeping too many irons in the fire and spreading myself too thin, assuming someone has decided they want to have a job like yours, what are the right questions to be asking? And while you're at it maybe you could answer a few of them as well.

The big question is do you *really* want a job like mine, if so why. I like my job because I play with lots of math and physics and it's the most geeky think that I've been able to find. If you want something different, then you probably don't want to do what I did. One thing that I like about my job is that a lot of it involves "figuring stuff out" and if you want something to give you instructions to do X and get Y, its probably not a good line of work.

Also, there is a limit to how much help that I can give you because history doesn't repeat. I stumbled into my job because of several historical events that simply will not repeat themselves. My story might be useful (either as a guide or as a warning), but I can't give you a cookbook, because reality just doesn't work that way. You could do exactly what I did, and you won't get the same result because its 2011 and not 2001, and even if I did exactly what I did, quantum randomness would probably insure that there would be a different outcome.

One thing that helped me a lot was to study the history of the Roman Empire and learn to read Latin. If you look at a lot of history, you see patterns, and patterns can help you figure out what to do next. It's worth reading Thucydides and the History of the Peloponnesian War. Once you read how empires rise and fall, you can look at the news and then figure out what happens next, and what you can and should do about it.

Also read Chomsky, Marx, Confucius, and Plato. They can provide more insight as to "what questions should I be asking" than I can.

 

[twofish-quant]

An example job posting, it tells you exactly what they are looking for. We are all pretty smart people we can figure out what degrees and training will give us these qualifications.

Job postings lie.

Except for security clearances and visa status, a job posting will *never* tell you what the employer is really looking for. The only thing that you can figure out from a job posting is that there maybe a job available, and sometimes that's not even true.

Two-fish, would you say with the multitude of MFE programs out there and the rising recognition of the "quant career field," (saturation) that the hardest part of becoming a quant is getting your foot in the door?

It's hard to get a job. It's just as hard to keep one.

The hard part is keeping sane. I've found reading absurdist literature helps a lot.

 

[twofish-quant]

One thing that I've found useful is reading Chomsky and Marxist critical theory. One thing that you quickly figure out is that one of the purposes of education is to turn you into a well behaved corporate cog. So what happens is that the educational system trains you into asking only certain questions and to think in a certain way. It's like a horse that has blinders in which you put a carrot in front.

Nothing wrong with that, as long as you get the carrot. However, you run into big problems once you figure out that there is no carrot. So rather than asking "what do I need to do to get a job" I think it's better to ask "why are there no jobs?"

 

[chiro]

One thing that I've found useful is reading Chomsky and Marxist critical theory. One thing that you quickly figure out is that one of the purposes of education is to turn you into a well behaved corporate cog. So what happens is that the educational system trains you into asking only certain questions and to think in a certain way. It's like a horse that has blinders in which you put a carrot in front.

Nothing wrong with that, as long as you get the carrot. However, you run into big problems once you figure out that there is no carrot. So rather than asking "what do I need to do to get a job" I think it's better to ask "why are there no jobs?"

Don't you see the irony though in this?

You always hear about the so called competitive global world and economy and how people are saying that we need better educated workers who will innovate, yet at the same time, there are education systems that enforce the exact opposite (high school and below)?

I think that although the pre-tertiary systems are like this, universities are a lot different in that they do provide more opportunities to think for oneself.

One thing that has stuck out is that there are some universities where innovation seems to be more nurtured than everywhere else. The examples that stick out are places like Stanford and MIT with technology related startups. Both of these universities have very good technology entrepreneurship basis (especially Stanford) and they also have good departments that advise on things like licensing and other legal issues. Stanford naturally is in an area of venture capitalists which also helps out a lot.

I remember you posting before about Xerox and that their one inhouse invention (it was either the photocopier or the laser printer) was all it took to fund the rest of their investigations.

To me the above is what a lot of people don't understand: they see the Google's and the Yahoo's and the Microsofts, but they forget about the Segways and all of the other things that didn't work out.

So with regard to there being "no carrot", the thing is that innovation sometimes requires the carrot to be somewhere else and that sometimes the horse and its jockey need to actively seek it, often in the hope that it even exists to begin with.

 

[Poopsilon]

Having a philosophy degree myself and turning from there to pure mathematics has made me reluctant to wax philosophic and generally disdainful of the entire practice as first conceived. Rigor pays the bills, and I intend to build my house out of human teeth.

Chomsky is a fine man, and maybe if anyone who actually mattered listened to him our species wouldn't be in the mess it's in, but they don't. I haven't gotten a chance to read much of the frankfurt school but I remember reading Lyotard and feeling happy with what little I understood.

You shouldn't worry about your advice being taken as dogma, at least not by me. It will get filed away in my brain with a hundred other sources to be combed over later as I attempt to build a picture clear enough to chart my path. When I ask for advice I'm not interested in having my worldview critiqued, I want to hear some real down to Earth all american practical thoughts on careers in finance. No cookbooks, no cookie cutters, just some simple straight forward advice that will ensure that I get a senior quant position at a top hedge fund. Starting salary I'm thinking 500k, what do you think?

 

[npc]

There are about 20 different jobs that qualify as "quant." It so happens that the work that I do *is* pretty close to physics. In graduate school, I spent 12 hours/day working on numerical code. Today, I do pretty much the same thing and crunch pretty much the same equations, and do the same sort of modelling.

True, it is similar, but it is still not physics and doesn't change my view that people should not see a physics PhD as a route to a finance job, if you are interested in finance then study finance. Only study physics if you are truly interested in it.

That's not the attraction for me. It helps, but you can make decent amounts of money working elsewhere.

Fair point, I should have probably elaborated my original point, people see this as a good career salary-wise, but there are better ways to make decent money then doing a Physics PhD with the goal of going into finance.

People are looking for a blue print on what is the best way to become a quant, but PhD in physics is not a simple answer. In our area there are 6 'quants', only one has a PhD, the rest have masters, backgrounds vary from chemistry, physics, math, computers and an engineer. The important thing is that we can all understand the maths and are able to use computers to a decent level, from c++ to easier things like SQL and Excel vba (I often think people overestimate the use of c++ in finance and underestimate the amount of time on less demanding tools like excel/SQL). But we are a small regional bank, the bigger investment houses may very well only hire PhD grad because there are enough of them looking for jobs.

Like getting any other job, I would say luck & experience are more important in getting a quant job then anything else.

 

[Poopsilon]

npc, could you talk a bit about what the high moments are for you in a good day at work, as well as what types of problems you are deriving your intellectual satisfaction from, and what types of math you are generally using. Thanks.

Edit: also whether you're in new york or not.

 

[twofish-quant]

No cookbooks, no cookie cutters, just some simple straight forward advice that will ensure that I get a senior quant position at a top hedge fund. Starting salary I'm thinking 500k, what do you think?

That's not possible.

I can imagine a society in which people can produce enough wealth so that most people make $100-$200K. Efficient capital investment. Technology growth. Yadda. Yadda. Yadda.

I can't imagine a society in which you generate $500K/person. Which means that in order to make $500K, you have to be "special" and the odds are that you (or I or anyone else) aren't.

If some offers me a job that pays $150K-$200K, I can see how that much value is generated by my work. If someone offers me a job that pays $500K, I don't see how that can work. And if someone offers me money whose source isn't immediately obvious, I get very nervous that I'm being taken for a ride. I don't have any moral objection to taking money that is offered, but it's a matter of not getting stabbed in the back.

This influences the advice that I give. If I tell 100 people to go into science and technology, I can imagine a world in which that generates enough wealth so that everyone makes $200K. I can't imagine how everyone ends up making $500K or $1M. Which means that I'm not offering "real cash" I'm just offering lottery tickets, and having been screwed by this myself, I'm not in the mood to screw over other people with this.

 

[twofish-quant]

True, it is similar, but it is still not physics and doesn't change my view that people should not see a physics PhD as a route to a finance job, if you are interested in finance then study finance.

But people can see finance as a route to a physics job. The stuff that I'm doing is basically physics with financial data. One should point out that there are hundreds of different jobs in finance, and 99% have nothing really to do with physics.

Fair point, I should have probably elaborated my original point, people see this as a good career salary-wise, but there are better ways to make decent money then doing a Physics PhD with the goal of going into finance.

No argument here. But there aren't too may better ways of doing finance to do something more or less resembling physics and get paid decent amounts of money. If your main goal is to do finance or make money, a physics Ph.D. makes no sense at all. If you want to do a physics Ph.D. and are worried about starving to death, that's another issue altogether.

In our area there are 6 'quants', only one has a PhD, the rest have masters, backgrounds vary from chemistry, physics, math, computers and an engineer.

My group is about half Ph.d.'s and half masters with lots of computer experience.

(I often think people overestimate the use of c++ in finance and underestimate the amount of time on less demanding tools like excel/SQL).

Depends on the job. Most of my programming is C++.

But we are a small regional bank, the bigger investment houses may very well only hire PhD grad because there are enough of them looking for jobs.

This is a problem since one big, big gotcha for physics Ph.D.'s in finance is that you have to live in NYC, London, HK, or Singapore. If you don't like NYC (and many people don't), it's going to be a problem.

 

[hatelove]

I'm considering dropping my interest in science completely by not going for a matsci eng degree and getting either a degree in statistics and a PhD in applied math, or a degree in applied math and a PhD in statistics. I'm not sure which route to take, though.

 

[Locrian]

I was looking for something cross between applied mathematics and statistics, and I've recently discovered something called "actuarial science..." This might be the best option I've found so far because I wanted to go into something like applied statistics (which I don't think exists) and actuarial science seems pretty close.

There are a few actuarial jobs that would fit this description, but I think most would not. You cannot be an actuary without having a background in statistics, but that doesn’t mean you’ll use it very often in most actuarial work. The reason actuaries are required to pass the actuarial exams is because the math we need for our work is really not covered well in traditional courses. You could try to land a more mathematically intensive actuarial job, but I don’t think it’s wise; you would be reducing your pool of potential employers enough that the probability of getting a job would be unacceptably low.

Actuarial jobs are ultimately business jobs where you must understand some concepts in probability and statistics; they are not mathematical jobs in business. For that reason I’m skeptical about whether this career is the one you’re looking for, given your posts in this one thread. However, if you want to learn more feel free to message me, start a new thread or update an old one on the subject and I’d be happy to discuss it more.

 

[deRham]

No cookbooks, no cookie cutters, just some simple straight forward advice that will ensure that I get a senior quant position at a top hedge fund. Starting salary I'm thinking 500k, what do you think?

You consider a strategy that will ensure a salary of half a million to be something other than a cookbook?

Someone said (after having lots of math, programming in general) a lot of luck and experience is what counts most. I am pretty sure that holds for many fields. I think there is a pretty big limit on what you can just forcibly train yourself to do in school, if measuring success by advancement and monetary compensation in the 'quant' area, as opposed to simple intellectual thrill.

 

[npc]

npc, could you talk a bit about what the high moments are for you in a good day at work, as well as what types of problems you are deriving your intellectual satisfaction from, and what types of math you are generally using. Thanks.

Edit: also whether you're in new york or not.

First off, I'm not in new york, my perspective of this type of work will be different to twofish. I work in a regional european bank outside UK.

The work we do will be different as, where in a large finance house in london/NY etc, they will tend to hire a quant to work in a specific area like equity, fx or interest, in the smaller houses the quant people will need a broader knowledge as they will be covering all products. The daily work ranges from coding up a finite differece or monte carlo model in c++ for a new product to discussing hedge accounting rules with accountants to talking to people about what the best interest rates are to use for pricing various products.

Intellectually the work can be both satisfying and frustrating. This is not like physics or science in general where you can objectively look for a 'correct' answer. In this field the correct answer can be quite subjective as the correct answer is often the one that traders&markets can do business at, as a result we have to work on the principle that most of your models will be incorrect within 2/3 year and need to be re-designed due to some market shifts. Some of the models & parameters can be quite arbitrary. It can all be quite a shift in thinking from regular science.


On a more philosophical note, over the years I have met many people in science/engineer who have a principled view that they will never work in the armament production/development industry, because they would no want their work hurting anyone. But I have met few people who have the same stand when it comes to working in finance, the reality is that the quantitative people in all financial institutions were at the forefront of the enablement of the current financial disaster in europe and beyond. While no war has started (yet) because of the financial mess, clearly there has been a lot of damage to people & society. I am not blaming the individuals (I am one of them), but it is sobering to think you are part of the machine...

Anyway, that my two cents...now back to work as a little cog in the big machine.