Understanding R

I’ve had harsh things to say about the R statistical package in the past.  However, for the past few weeks I’ve been using it a lot and I have to say, it has grown on me.

I’ve always realised that it’s an advanced and powerful statistical work environment (based of course on the S language/environment).  Four things have caused me problems: my own inadequate understanding of anything more than basic statistics, the lack of clearly-explained examples, the difficult-to-learn syntax and the unhelpful documentation and mailing lists, which give an unfriendly impression of "by statisticians for statisticians".

Well, it’s like anything else.  If you invest some time and use Google effectively, you’ll find some useful resources and get your head around it eventually.  Many maths and stats departments have posted useful material online (such
as these code snippets
) and there are some good tutorials and examples too – even geography undergraduates learn more than me!.  I may post some more links and PDFs at some stage.

There’s a heap of bioinformatic-related techniques that are implemented in R packages – SVMs, neural networks and so on, so I think it’s well worth the effort.