Free Software

sl_0.0-5.tar.gz and sl_0.0-5.zip are unix and windows R packages implelenting the examples in the paper "Statistical Inference for noisy nonlinear ecological dynamic systems" Nature 466(26):1102-1104. (Originally Nature corrupted these en route to putting them on their web-site.)

Soap film smoothers are included in R package mgcv from version 1.7-21. The code there is based on sparse matrix methods for solving the defining PDEs and is somewhat faster than the code here. soap_0.1-5.tar.gz implements soap film smoothing based on multigrid PDE solution, but is no longer maintained. You'll need mgcv 1.5-6 or above for this to work. See the examples in help("soap-package") to get started. The paper describing this is Wood, S.N., M.V. Bravington and S.L. Hedley (2008) "Soap film smoothing" JRSSB 70(5).

Solv95 - A delay differential equation solver for Windows operating systems. This is open source, and comes with examples and documentation. All you need to use it is a C/C++ compiler: gcc (which is free) works fine, and links to a suitable port are provided. This is fairly mature and reasonably well tested (however I've received reports that the windows interface no longer compiles properly on some platforms).

mgcv - an R package (also usable with S-PLUS) for multiple smoothing parameter selection using GCV for generalized ridge regression problems. Also includes C source that can be used independent of R or S-PLUS.

DDEfit A Windows package that allows you to fit dynamic models specified as delay differential equations, differential equations or difference equations to data (models may be partially specified). The model is specified by editing a template file written in C, which is then compiled and linked to the libraries containing the numerical, statistical and graphical routines required. You need a C/C++ compiler to use this (gcc works). This package is more ambitious in scope than the last two, so is more likely to have some glitches.

posum 2 Mortality estimation software based on the population surface method. The method is implemented as an open source R package, but relies on a rather old version: will try to fix soon.

Please let me know if you have any problems installing or using these: they are intended to work, so I'd like to know if they don't! My email is: s.wood@bath.ac.uk