DDEfit 0.504

Posted 18/3/2003.
What's new and what's fixed in this version.
This page contains the files needed to use the DDEfit program. This is a Windows 9*/NT based package for fitting (delay) differential equations and discrete time models to data. It is written for the purpose of fitting ecological process models. The software is provided under the following conditions:
Please read this before downloading: GNU General Public License .
You will need a GNU, Borland or Microsoft C/C++ compiler capable of producing Win9*/NT 32 bit multi-threaded GUI applications. I used Microsoft Visual C/C++ 6 and an excellent port of gcc (see below for link) Please drop me an email if you download and use this package. My address is simon@stats.gla.ac.uk
The reference for methods implemented in this code is Wood, S.N. (2001) "Partially Specified Ecological Models" in press Ecological Monographs 71. See preprints page for a pre-print. The basic concept was first suggested in the discussion of: Wood, S.N. (1994) "Obtaining birth and mortality patterns from structured population trajectories" Ecological Monographs 64(1):23-44.
DDEfit is open source. The package contains a fair number of files, so they are provided in a zipped archive: If you want to use gcc you need to have a version of the compiler capable of compiling to 32 bit native windows applications. Several are available. The one that I have used is EGCS Mingw32 Native.. It's stand alone and comes with gcc , a resource compiler windres, and a debugger gdb. To learn about GNU C for Win32 (including other versions of the compiler), try the gcc for Win32 page.


Thanks to Roger Nisbet, Bruce Kendall, Doug Donalson and Andy Paul for comments, bug reports and suggestions.


Please email me with bug reports and I'll try and fix them. To facilitate bug fixing I need to be able to reproduce the problem, so it's really helpful to have the model definition file, data file and weight file (also any parameter file that might be needed) that causes the problem, along with a description of what to do to cause it to happen. Any models and data sent for this purpose will be treated as confidential and used only for debugging! (Please let me know about non- reproducible bugs as well, so that I can keep an eye out for possible causes of them).
Home | Back