A plan to train myself as a quant

[charlieLuo]

I am planning to move to the field of financial engineering. I had classes in calculus and differential equations, probability/statistics/regression, discrete math, stochastics, and simulation. That's basically my math background. Some classes were very remote to me now.

Here is my plan to train my self the BASICS of a quant (I'd appreciate suggestions and information for me).

1 month, take GRE math. just to refresh my basic math
1.5 months, Stochastic Calculus, by Steven Shreve
1.5 months, Numerical methods, by Paolo Brandimatre
2 months, Options, Futures, Other Derivatives, by John Hull

I will do this one by one. So total is about 6months for me. I might slow down if I find it is too difficult for me. For now, it is my plan.

Any suggestions? Especially if you think I am missing some important building blocks.

 

[MikeK]

Here is my plan to train my self the BASICS of a quant (I'd appreciate suggestions and information for me).

I see from your intro that you want to train yourself to be a quant. I have no idea how easy or difficult that is. Do you have any advanced degree in engineering or computer science or anything like that? Does your current job give you an opportunity to move to a more quant-oriented role?

I don't know what is possible, but here's a good thread on books to read:
Master list of useful Quantitative Finance books - QuantNetwork - Financial Engineering Forum

However, this is just a general list of good books...not a "how to" for people that want to train themselves to become a quant.

 

[charlieLuo]

thanks very much, Mikek. I do have advanced degree in computer science, statistics, some business.

I am not sure my goal between a FE program / working as a quant.

Will I be able to perform a entry level quant's job, if I learn those books as I listed in the original post?

If so, I probably want to work and get into some FE program part time.

 

[John]

let me propose an alternative plan: you can do this side by side with your other fundamental study ( the way you had it laid out). Pick an interesting project in financial engineering, say pricing of an exotic equity option like himalayan option. While you tackle this project, you will surely hit a roadblock, like something lacking in your background that you need to get over that hump. I found out in my experience that I learn better when I have a concrete project in mind.

Once you have mastered equity derivative, you move on to the following

1. interest and credit curve modeling
2. short rate modeling and interest derivative
3. credit derivative

 

[MikeK]

Will I be able to perform a entry level quant's job, if I learn those books as I listed in the original post?

Personally, I think you'd be better off getting more formal training in an FE program. While those books are probably all good, the FE programs also focus on applying what you learn to real-life trading situations, etc. Many FE programs are continuously modifying their curriculum to keep up with the fast-changing industry.

You could certainly brush up with those books and apply for some entry-level quant jobs. That wouldn't hurt. If you get some bites, great. If not, you can re-evaluate your plan to enter that career path. If you have a good C++ background, some stats, etc, you may get some interviews. I don't really know. If I was a hiring manager, and you were applying for a portfolio management job, I'd probably give you at least a phone interview. Depends how strong your finance/business is.

 

[charlieLuo]

let me propose an alternative plan: you can do this side by side with your other fundamental study ( the way you had it laid out). Pick an interesting project in financial engineering, say pricing of an exotic equity option like himalayan option. ....

I like your idea very much. I am a bit embarrased to tell you that, I do not have any idea what are potential interesting porjects, neither where to find them.

Would you point out a place (a paper/book/website ... etc), ideally comes with some sorta list?

Thanks. And I sincerely appreciate your help!

 

[John]

let me propose an alternative plan: you can do this side by side with your other fundamental study ( the way you had it laid out). Pick an interesting project in financial engineering, say pricing of an exotic equity option like himalayan option. ....

I like your idea very much. I am a bit embarrased to tell you that, I do not have any idea what are potential interesting porjects, neither where to find them.

Would you point out a place (a paper/book/website ... etc), ideally comes with some sorta list?

Thanks. And I sincerely appreciate your help!

Amazon.com: Financial Engineering and Computation: Principles, Mathematics, and Algorithms: Yuh-Dauh Lyuu: Books

I like this book and I think it would give you a good overview of the products and models in FE, plus it provides pseudocodes for most of the models. The explanations are a bit sketchy though, so you need to google or rely on other books for additional explanation.

 

[charlieLuo]

Personally, I think you'd be better off getting more formal training in an FE program. While those books are probably all good, the FE programs also focus on applying what you learn to real-life trading situations, etc. Many FE programs are continuously modifying their curriculum to keep up with the fast-changing industry.

.... If you have a good C++ background, some stats, etc, you may get some interviews..

Thanks much, Mikek. I see two valuable items I can earn from a FE program,
(1) knowldge and skills
(2) brand and industry connections
(3) others (?)

Any good program would you recommend? Any differences between schools at NYC, or Chicago, or S.F., in terms of industry connections?

 

[Andy Nguyen]

Reading posts here just show you how important that is to get the first quant job. People with PhD and/or years of experience on/off Wall Street have difficulty to knock that first hurdle off.
Getting the first job in the industry is one of the main reasons why people join MFE programs (and willing to pay through the nose for it).
Programs in NYC obviously have many advantages in locations/connections/job leads/etc but it also depends on how hard the director of the program willing to work to help his students.

 

[charlieLuo]

As you Mikek pointed out, it helps if I can set a clear goal, before I spend a year or two on FE training.

For now, assume I am equally interested in all quant jobs: a quant developer/a statistical arbitrage quant / a quant trader

So here is my background, BS in computer science, MS in statistics, and some advanced degree in business.

mathematics: I had stochastics and simulation before. after a year or two's FE training, I am sure I will learn alot more, but will not be as good as a decent MIT math master. I can make myself as close as possible to them if I focus on this.

statistics/econometrics: I consider my statistics background as "decent". with some quick prep, I can be a decent statistician.

computer programming: I was a C programmer. So I guess I need to pick up C++, and be good at it. Right?

finance: my finance background probably is similar to a freshout MBA who does not have much working experience, i.e., some knowledge of corporate finance, portfolio.

Other business& econ: I am an expert in project and product development management. and have comprehensive knowledge of general business/econ topics. I certaintly see a potentail to look into some specific areas, if I position myself as a trader.

So, assume you are a recruiter or a heandhunter, which of the following do you think I can reasonably easy to "pimp" myslef after a year or two's FE training: as a quant developer, or a statistical arbitrage quant, or a quant trader?


Have a great weekend!

 

[MikeK]

mathematics: I had stochastics and simulation before. after a year or two's FE training, I am sure I will learn alot more, but will not be as good as a decent MIT math master. I can make myself as close as possible to them if I focus on this.

By "MIT math master", do you mean like the dude from "Good Will Hunting"? If so, then I say you could be a trader. I always imagined Matt Damon as a trader, where he keeps yelling "How 'bout them apples" into his turret.

 

[charlieLuo]

oh, well. "Good Will Hunting" dude is way more different from what I called "decent". anyhow.
Interests wise , I prefer a quant trader > a statistical arbitrage quant > a quant developer.
I guess most fellas here have similar preference. I might be wrong though. Realistically, I do not see myself as a quant trader in 3 years, but ultimatelt I will be there. So only two left, which I am not able to decide....hope I can hear some opinions from you guys here.

 

[MikeK]

Interests wise , I prefer a quant trader > a statistical arbitrage quant > a quant developer.
I guess most fellas here have similar preference. I might be wrong though. Realistically, I do not see myself as a quant trader in 3 years, but ultimatelt I will be there. So only two left, which I am not able to decide....hope I can hear some opinions from you guys here.

Are you familiar with the roles of each?

 

[bigbadwolf]

1.5 months, Stochastic Calculus, by Steven Shreve
1.5 months, Numerical methods, by Paolo Brandimatre
2 months, Options, Futures, Other Derivatives, by John Hull

I will do this one by one. So total is about 6months for me. I might slow down if I find it is too difficult for me. For now, it is my plan.

Any suggestions? Especially if you think I am missing some important building blocks.

1.5 months for Shreve and 2 months for Hull? The 2 volumes of Shreve will take you more time. Probably at least four months if you do nothing else. Hull, on the other hand, reads like a novel, and if you can do Shreve in four months (or even six), Hull will take you a small fraction of that time. If I had to suggest a single book to start with, it would be Benninga's "Financial Modeling using Excel." Then work through Hull. Then if you think Hull's explanation of stochastic is too heuristic, go to Shreve Vol.2 (but you might want to bypass Shreve altogether, to concentrate more on computing and numerical analysis and simulation). Another excellent place to start is Wilmott's "Introduction to Quantitative Finance."

 

[alain]

1.5 months, Stochastic Calculus, by Steven Shreve

Good Luck!!

 

[Yuriy]

I agree with Alain

1.5 months for Shreve's Stochastic Calculus is not just not enough. I think not many people can even handle that book on their own. Simple reading is not enough, you need to be able to solve most of the problems. But how are you going to know whether your solutions are correct.

 

[charlieLuo]

Thanks for the heads-up, guys. :tiphat:

Right now, I focus on GRE math test on April 12th. I started late. So far about 22 hours' work. I reviewed pre-calculus and single variable calc, and have solved couple hundred exercises. It is fine so far. I will see how much I can get on the test, though I have a day job and some travel this month.

I find GRE math is a good way to polish my rusty math. Good news is that I find myself still enjoy math.

After the April GRE math. I will revise my plan for this summer. And I will definately report my progress and collect advices from you guys.