which is more useful in the real world? Python or C++...

[diwei shi]

I have no idea of this and one director of a Top 10 quant program told me they are equally useful.
Just wanna see more opinions from people in the field.

 

[volmacro]

Someone familiar C++ can probably pick up basic Python fairly quickly. Someone familiar with Python would probably take a bit of time to pick up C++. If you're starting from scratch, Python is probably the one that will give you most bang for your buck -- you can do a lot in a short amount of learning, and almost anything you cannot figure out, you can find out on stackoverflow. My impression is that more and more places are using Python, including many hedge funds now. C++ would probably remain in place for high frequency trading, but more slower frequency systematic trading is being run off Python now.

 

[pingu]

I have no idea of this and one director of a Top 10 quant program told me they are equally useful.
Just wanna see more opinions from people in the field.

They are useful in different ways.

IMHO, Python will help you solve problems faster. You will only use C++ when you need to but that will be rarely.

 

[Rafael F]

It depends on what you want to do. For daily activities, python. For more performance-needy systems, C++.

 

[Daniel Duffy]

It is possible to answer this question from a million perspectives. C++ is a bit like Latin (but not as old!) in the sense that if you know it all the other languages are easy to learn. A serious C++ developer can learn Python in a few days/weeks/months.

I get the feeling that Python is the new Matlab in a sense.

The language used is what you are requested to use.

Don't forget C#.

 

[Rafael F]

It is possible to answer this question from a million perspectives. C++ is a bit like Latin (but not as old!) in the sense that if you know it all the other languages are easy to learn. A serious C++ developer can learn Python in a few days/weeks/months.

I get the feeling that Python is the new Matlab in a sense.

careful with "new matlab" stuff. Python's power is far bigger than doing math operations..

 

[Daniel Duffy]

careful with "new matlab" stuff. Python's power is far bigger than doing math operations..

Sure My remark was not based solely on what the language can or cannot do.
Notice I said "in a sense".

OK, what does Python have that Matlab does not? That's a good question.

 

[pingu]

Sure My remark was not based solely on what the language can or cannot do.
Notice I said "in a sense".

OK, what does Python have that Matlab does not? That's a good question.

For one, it's free.

But don't steer the conversation to Matlab. The OP asked about Python and C++.

As of today, I don't turn to C++ unless I really need to squeeze some really really efficient thing that Python doesn't provide. I don't code for a living, I solve financial problems and Python provides me with quick solutions (R does the same for me).

 

[Rafael F]

OK, what does Python have that Matlab does not? That's a good question.

I answered but as pingu said, thats not the point of the question. We can discuss elsehwere !

 

[Daniel Duffy]

I answered but as pingu said, thats not the point of the question. We can discuss elsehwere !

Relax; I only mentioned Matlab twice

BTW we've already had a C++ v Python thread on QN already. These discussions tend to become quite boring and very non-quantitative, like Yankees v Socks, Ajax v Feyenoord.

some ideas on objective metrics

1. Efficiency
2. Support for OOP, Generics, functional style
3. Stability and reliability
4. Maintainability of code
5. Integration with legacy system
6. Usability + productivity.
7. The kind of application to be used (PDE/FDM, Monte Carlo, CAD, hedging, Games!)
8. Which languages do your team use?

 

[pingu]

Relax; I only mentioned Matlab twice

BTW we've already had a C++ v Python thread on QN already. These discussions tend to become quite boring and very non-quantitative, like Yankees v Socks, Ajax v Feyenoord.

some ideas on objective metrics

1. Efficiency
2. Support for OOP, Generics, functional style
3. Stability and reliability
4. Maintainability of code
5. Integration with legacy system
6. Usability + productivity.
7. The kind of application to be used (PDE/FDM, Monte Carlo, CAD, hedging, Games!)
8. Which languages do your team use?

most of these are IT topics. There is some IT guy which are going to solve those for me. I want to do some quant analytics and I'm not concerned with any of those. What can help me figure out something? Nowadays, usually Python and/or R allow me to do that.

 

[Daniel Duffy]

most of these are IT topics. There is some IT guy which are going to solve those for me. I want to do some quant analytics and I'm not concerned with any of those. What can help me figure out something? Nowadays, usually Python and/or R allow me to do that.

So, the fact that these are useful to you personally answers OP's question?

It sounds somewhat idiosyncratic to be honest.

 

[yashyadav]

most of these are IT topics. There is some IT guy which are going to solve those for me. I want to do some quant analytics and I'm not concerned with any of those. What can help me figure out something? Nowadays, usually Python and/or R allow me to do that.

I feel this is applicable only if you are doing POC sort of analysis and never going to code yourself and you just present proofs on paper. Someone else do the job of coding this, in to actual.
Else, without any perception on these things you might just write "A CODE" but not the "industry usable code".

 

[pingu]

So, the fact that these are useful to you personally answers OP's question?

It sounds somewhat idiosyncratic to be honest.

the OP asked for most useful language. I gave him/her two choices: Python and R. (neither of them is C++).

at the end of the day, they are going to use whatever they feel comfortable with... probably Excel.

 

[pingu]

I feel this is applicable only if you are doing POC sort of analysis and never going to code yourself and you just present proofs on paper. Someone else do the job of coding this, in to actual.
Else, without any perception on these things you might just write "A CODE" but not the "industry usable code".

??

 

[Daniel Duffy]

yashyadav's question is pretty clear to me.

 

[Daniel Duffy]

the OP asked for most useful language. I gave him/her two choices: Python and R. (neither of them is C++).

at the end of the day, they are going to use whatever they feel comfortable with... probably Excel.

Actually, the question was C++ or Python.
But I don't see any motivation for the answer. Pros, cons that kind of stuff is lacking.

This thread is in 1st gear.

 

[pingu]

I have no idea of this and one director of a Top 10 quant program told me they are equally useful.
Just wanna see more opinions from people in the field.

what kind of problem are you trying to solve? Do you have a concrete example? That is probably a better question.

 

[TehRaio]

I get the feeling that Python is the new Matlab in a sense.

woah, where does that leave matlab? cause I'm really in that camp. it seems to be shunned by a lot of people -_-

 

[volmacro]

woah, where does that leave matlab? cause I'm really in that camp. it seems to be shunned by a lot of people -_-

MATLAB has 2 problems I think:
1. cost, it is not something you can just trial like you can with Python
2. its numerical data focus, doesn't work well with strings, etc, so it almost always needs to be supplemented