Importance of programming

[nitin]

Hi.. i am Nitin and looking forward to pursue MFE this coming fall (2009) .. I have been doing some research about MFE program and the career as such by myself and I found "programming" to be a core activity for any quant professional...

My doubt:

1) What depth of programming is necessary for doing the masters and pursuing the career in quant finance ( good enough or extremely good or GOD at programming)

2) Is it easy to pick up the skill in a few months of practice before and after joining the program

3) I am a computer science student myself.. but never indulged in serious programming before.. will 2-3 months be enough for me to excel at C#, MATLAB etc ..


Thanks

Nitin

 

[IlyaKEightSix]

Depends what kind of quant you want to be. If you want to write the actual software, then a ton.

Not sure how much the others.

 

[bigbadwolf]

1) What depth of programming is necessary for doing the masters and pursuing the career in quant finance ( good enough or extremely good or GOD at programming)

2) Is it easy to pick up the skill in a few months of practice before and after joining the program

3) I am a computer science student myself.. but never indulged in serious programming before.. will 2-3 months be enough for me to excel at C#, MATLAB etc ..

You need an aptitude and taste for scientific programming, which means the area where numerical analysis meets computer code. So an aptitude and taste for algorithms and for clean, tight, robust code. What you're ideally aiming for is not some nebulous "mastery" of C# or MATLAB, but how to use them in applications such as, for example, simulations or numerical solutions to PDEs. And if you try to learn C# and MATLAB with this kind of focus, your study will be more motivated and you will be able to sift what is relevant to your needs and what is not.

 

[buddyforall]

It is said that quants do 70% programming.I asked a smiliar question to someone with a MFE degree from Oxford who presently works for a big I bank.

The reply I got was this


"i am a quant developer. i work in a well-respected ib in a large group and most people around me have phds - guess what ... everybody is coding nearly all of the time and even senior people are debugging stuff.

if you are not into coding & debugging, i would not go in the "pure" quant direction. the really interesting stuff tends to be done by the seniors (unless you work maybe in a smaller place where the situation is less evolved). you could look at strategist positions (who come up with trading ideas) or trading, structuring. "

 

[Stefan Zota]

You need an aptitude and taste for scientific programming, which means the area where numerical analysis meets computer code. So an aptitude and taste for algorithms and for clean, tight, robust code. What you're ideally aiming for is not some nebulous "mastery" of C# or MATLAB, but how to use them in applications such as, for example, simulations or numerical solutions to PDEs. And if you try to learn C# and MATLAB with this kind of focus, your study will be more motivated and you will be able to sift what is relevant to your needs and what is not.

Correct, this may be rephrased a bit and put on a wiki as a general response to this question.

There are many areas to implementation. It can be pure software design, it can be operating systems programming, it can be numerical programming, it can graphics, it can be networking programming, it can be parallel programming etc. Each of them require different backgrounds, they overlap however expert level knowledge is not needed in all. For a quant, I would choose: numerical programming (conversion from theoretical algorithm to implementation) and software design (a side of "clean, tight, robust code").

 

[doug reich]

Correct, this may be rephrased a bit and put on a wiki as a general response to this question.

Agreed, I added it to the FAQ; see my signature.