Tools for Computational Economics

Programs

I believe that open and transparent science should rely mainly on open-source software. This is why I carry out most of my work with programs that are open-source. For the computational part of my research I rely on Python and R, two great programming languages. For version control I rely on Git and most of my papers are written in LaTex. Because security is also important I use OpenPGP (here is my public key).

R

R is a great prgramming language. Its main area of application is statistics, but its not limited to that. As it is the case for most open-source programs, its fantastic community is almost always able and willing to help you with your problems. R is also extremely powerful when it comes to data visualization. The big disadvantag is a lack of a good graphical interface, although R Studio partly invalidates this criticism. The main advantage for students is that R makes one understand what is really going on in statistics. I therefore think that all students working with statistics or econometrics should have a basic understanding of R. It makes you more independent. Also, if you know R, its easier to learn the common GUI alternatives such as STATA or SPSS. And most importantly, its free and will allow researchers all over the world to study the important questions of our time.

I am teaching an R course at the University of Bremen. You can get all the material here.

Python

Python is an easy and very general programming language. It works perfectly together with R, but has a wider scope of applicability. Especially when it comes to agent-based simulations I consider it to be more intuitive than R. Yet, I think it is a good decision to learn both languages: R is more common and you will have more people to work with. Python allows you to do any advanced scientific programming of interest in the social sciences, including agent-based modeling and network analysis. Besides, Python is a lot of fun and can help you out with a lot of everyday-problems. In fact, this whole webasite was created using Python!

Latex

One of the best decisions in my undergraduate studies was to learn LaTex. Although the first seminar paper will be painfull, Latex will save an increadible amount of time in the medium run. Also, documents are much nicer than typical word documents and LaTex is platform independent and free. It also works well with R (through the fantastic knitr package) and most journals offer you templates that save a lot of time when it comes to final formatting.

Git

Git is a version control system that helps you to organize your own work and to cooperate with others.

OpenPGP

PGP (which stands for “Pretty Good Privacy”) is an encryption software that helps you to sign, encrypt, and decrypt files. You can get my public key here).

Back to main page.