With regards to writing fast and efficient code, it is a matter of programming ideas and of libraries, not of language.
C++ might be one of many “purest” object oriented language utilized by quants at present. It is received various ideas like a number of inheritance that may be extremely highly effective. However I’ve seen folks utilizing it as a procedural language as nicely, and that is probably not useful.
Python is way from only a nice prototyping language. Its key energy is its intensive and highly effective libraries used for numerical and statistical calculations. For instance, I can generate a VaR timeseries in however a single line.
As with C++, you are able to do horrible issues to it – programmers from an IT background hold iterating over matrices as an alternative of utilizing vectorized capabilities. Used correctly, it isn’t materially slower than C++.
Neither language comes with out dangers both. For Python it is a lack of robust typing whereas C++ has the distinctive functionality of manufacturing reminiscence leaks.
What makes Python stand out to me nonetheless is the flexibility to make use of the identical language throught the quantitative course of, all the way in which from analysis to manufacturing. If I can get my palms on some top-notch C++ quants, nice. If I’ve to decide on between the chance of weak typing and IT guys recoding fashions in C++ and not using a clue what they’re doing, I’ll select weak typing any day.
Have a confidential story, tip, or remark you’d prefer to share? Contact: [email protected] within the first occasion.
Bear with us in the event you go away a remark on the backside of this text: all our feedback are moderated by human beings. Generally these people is perhaps asleep, or away from their desks, so it might take some time to your remark to seem. Finally it can – until it’s offensive or libelous (wherein case it received’t.)
Picture by cottonbro studio: https://www.pexels.com/picture/hands-typing-on-a-laptop-keyboard-5483149/
