Ask HN: Econometrics, Bioinformatics, Game Theory. Python or R? | Hacker News Reader

Ask HN: Econometrics, Bioinformatics, Game Theory. Python or R?

4 pointsbarry-cotter5 comments

Come next October I'll be studying Econometrics, and that seems as good a reason as any to finally take the plunge and learn to programme.

Python or R?

I'm under the impression that Python is a lot easier for a total novice to get into, and since its designed to be a general programming language rather than a statistical package that's Turing complete I'd be able to move to simulating games with the greatest of ease.

Also, IAUTI that bioinformatics really, really likes NumPy and SciPy and if I can grok the statistics behind econometrics it'd be a sane amount of work to learn another area I'm really interested in.

Recommendations, tutorials, books please.

View on HN

I'd recommend Sage. It is a math-oriented Python environment that includes many of the popular open source math packages. It also has a worksheet mode that lets you save an interactive session, which is something R can do but not python.

In fact, Sage includes R so if you really want to you can use R functions from within python (unfortunately, it's a little messy).

http://www.sagemath.org/

I would recommend Python as a first language. There are tons of users (which is very nice when you run into problems), and it's useful for all sorts of problem domains.

http://diveintopython.org/ is a good place to start.

(Or, if you aren't going to start learning in the immediate future, this is looking pretty promising:

http://diveintopython3.org/ )

O'Reilly's "Learning Python" seems pretty decent too: http://oreilly.com/catalog/9780596513986/

I really liked How to think Like A Computer Scientist http://openbookproject.net/thinkCSpy/index.xhtml

A friendly way into programming in general and uses Python.

Also, there are ways to connect R and Python.

As a programming language Python would be more suitable, 'cos it's more generic.

When it comes to Econometrics or any other statistical analysis, you would probably call built-in/library/module functions to calculate things like Gini, Theil.

Most of these functions are available for python or even R. So you can learn the basic syntax for both, and figure out what is the best way to do what you need, as you go along.

my friend told me that econometrics is his hardest phd qual. iirc he mentioned statistics, time series, and stochastics (very heavy maths) ... to me those seem to be best handled with statistical package like R

my bio friend told me about string matching, she used perl. i'm not sure if R can do pcre regex pattern matching.

so yeah, you may end up using R and python. if i have to decide, i'll use pyton, because it's more flexible

Reply on news.ycombinator.com