Best course/site to learn Python for risk managers

Hi,

Anyone know some respectable Python course for risk managers?
I know the basics of Python but I’d like to do some specialised exercises/courses designed for risk managers.

Thanks a bunch.

if i may ask, why not R? it’s a powerful and intuitive programming language designed specifically to tackle statistical problems, including risk management. you can find community-written packages to handle pretty much anything. there is an online course at Coursera which uses R:


note: coursera also offers a python course, but just the basics.

if you really want to use python, you might benefit more from reading a book instead, such as
Bayesian Analysis with Python: Introduction to statistical modeling and probabilistic programming using PyMC3 and ArviZ.

R is next on my list.
I find Python to be more “all-around” programming language which applies more to other areas of finance, not just risk management. For now, I think it has more bang for your buck. But I might be wrong, I’m in no way shape or form an expert in coding.

there is a never ending debate on whether R or python is better. for the purposes of working with data, i would say that the winner is R.
Python has a vast library, but when it comes to anything useful for data science/ data analysis, well…

comparison between python and R:

i strongly disagree with his veredict on Elegance though.
R makes is extremely easy for financial analysts. for example, to find the black scholes price, greeks, or the price of many exotic options, all you have to do is add a package and call a function with the proper parameters. two lines of easily readable code. R can also produce elegant graphics with ggplot.

Python is powerful, dont get me wrong. but you may have to re-invent the wheel many times over when analysing data. with R, all you have to do is add a community-writen package and off you go.

I use both as each has its strengths and weaknesses. I tend to do all my connections in pythons (ie api hookups) but it’s hard to beat the intuitive modeling that Rstudio is capable of so a lot of my analysis and graphing is in r. Early on I used codecademy and datacamp to learn python (it’s basic stuff but it’ll give you a foundation to build off of) and i think you can likely self-teach R.

I honestly don’t know how python would be different for risk manager than any other user.If you are reffering to some risk specific library,I am not aware of any sites that teach those libraries. However if you are just looking to learn Python with some NumPy and Pandas then maybe look into dataquest.io .They charge around $50 a month but teach you most of the basic “data science” stuff you need to know.
If you are looking for anything specifc feel free to ask me.

R intuitive? What a load of propaganda. Fake news at its finest.

Learn python and leave the r dinosaurs to their mail order dvds from Netflix and flip phones.

Best way to learn is by doing. If you want data skills do the pandas series on YouTube by Kevin Markham of data school.

R benefit is it is built by statisticians so it has great defaults for stats stuff. R con is that… It’s built by statisticians and sucks as a language.

Get with the times or get left behind. I hear they need some cobol programmers in the northeast

by comparison,

R was released over a decade after Python, which existed long before netflix was even founded. Now that’s old. Unlike other generic languages, R was developed by people who actually use data in their job to make their lives easier.

Okay dinosaur, please give me the usage weighted age. I’ll wait for you to spin your outdated, academic utopia. Speaking on Netflix…

What the ■■■■? Why didn’t they build their tech on R? Oh yea… because excuse excuse dinosaur!

Google, Facebook, Twitter, Uber, AirBnB, IBM, The New York Times, Firefox, JP Morgan, Morgan Stanley, HSBC among many, many others use R to handle data.

IMG_1737

it should be noted the obvious:
R is mostly meant for data analysis. if you want to build stuff, you are better off using a generic programming language instead. (there is also a fierce debate of which is better here. ps C/C++/C# is better in this case, but it does have a very steep learning curve)

haha this gave me a good laugh - why all the hatred towards R? It shouldn’t be one or the other - in fact I’ve taking a liking to js and will likely dabble in rails later on in the year.

I’m with you, i like python more and tend to use it more often than not but when I’m running regressions R is far easier to get an output, make transformations, whatever you need.

what’s everyone use for a code editor?

back when i was learning, i would use ubuntu terminal + gvim. working through terminal is imnho faster and more efficient, at least for school work. ubuntu, because windows command sucks for coding (again, my opinion)

Just trolling this old man.

I use VS Code. I haven’t modeled much in R yet because I’m studying my Fortran, but regressions are just two or so lines in Python using statsmodel. I think things have gotten much easier. R has packages built out that don’t exist in Python so I sometimes use it.

I heard someone say python is the second best programming language and I agree. It’s not #1 in anything but it’s the only one pretty strong in lots of categories.

Our system uses Javascript to customize it. Can’t believe you like that language

haha, well keep at it then man. regressions in r are one line - no library need - just saying.

I know nothing about fortran so ill need to check it out next and i’ve found quite a few functionalities in js that i wish python had but i’m just getting going in learning so the reason i like it is probably due to the fact I havent covered more complex areas yet.

I was joking about fortran :slight_smile: I have been learning Bayesian statistics and will likely use R for part of my learning. Like I said, its benefit is it is built by statisticians. Better defaults and such.

I haven’t done anything complex in Javascript. Just some financial models.