Are you also sometimes lost in the statistical methods labyrinth? Recently, I wanted to find a simple way to construct probability density curves (think of histograms except fitted by a nice curve). I leant back in my chair with the thought that this should be an easy one. A 'one-hour-max-job'. Guess what happened several hours later?
... I was still sitting in my chair, flipping through pages of technical descriptions about different ways of how to fit density curves. Should I use kernel smoothers with adaptive nearest-neighbor or generalized nearest neighbor bandwidth selection? Or would I be better off with a local likelihood approach or a penalized likelihood approach? Argh!
The Computational Methods Special Group of the BES must have heard our collective sighs. It is preparing a 'Field Guide on Computational Methods' - an internet platform where users can read and post short and friendly descriptions of computational methods. During a workshop at the meeting, we have discussed ways of how to realize it and finetune it to users' needs. For example, we came up with the idea to feature pictures and videos.
The Field Guide is a cool idea and I am looking forward to seeing it online.
The Field Guide is a cool idea and I am looking forward to seeing it online.