Comparisons of System Dynamic Simulation...
Posted: Mon Apr 28, 1997 11:52 am
We mention the combined use of statistical modeling with dynamic modeling
in our books: Dynamic Modeling in Biology, - - in Economics, Springer
Verlag, May 1997 (Hannon, Ruth)
One needs statistical modeling to represent the input of information to the
dynamic model in those areas that you feel you cannont represent them by
extension of the model structure--either because it is thought to a small
influence or it would be too costly to do so. And the statistical modeling
approach is useful on the final end as one tunes the parameters of the
dynamic model either to fit known data or to optimize the model output. One
of the major under-investigated areas in modeling today is the use of
statistical methods to determine the sensititvity of the variations in
parameter combinations in very large dynamic models. Some of our cellular
models for examples, have thousands of parameters whose variations and
combinations of variations may consort to cause real anomolies in the main
model output.
See: http://blizzard.gis.uiuc.edu/htmldocs/TES.html
Bruce Hannon, Jubilee Professor
Liberal Arts and Sciences
Geog/NCSA
220 Davenport Hall, MC 150
University of Illinois
Urbana, IL 61801
reply to: b-hannon@uiuc.edu
http://www.gis.uiuc.edu
217 333-0348 office
217 244 1785 fax
in our books: Dynamic Modeling in Biology, - - in Economics, Springer
Verlag, May 1997 (Hannon, Ruth)
One needs statistical modeling to represent the input of information to the
dynamic model in those areas that you feel you cannont represent them by
extension of the model structure--either because it is thought to a small
influence or it would be too costly to do so. And the statistical modeling
approach is useful on the final end as one tunes the parameters of the
dynamic model either to fit known data or to optimize the model output. One
of the major under-investigated areas in modeling today is the use of
statistical methods to determine the sensititvity of the variations in
parameter combinations in very large dynamic models. Some of our cellular
models for examples, have thousands of parameters whose variations and
combinations of variations may consort to cause real anomolies in the main
model output.
See: http://blizzard.gis.uiuc.edu/htmldocs/TES.html
Bruce Hannon, Jubilee Professor
Liberal Arts and Sciences
Geog/NCSA
220 Davenport Hall, MC 150
University of Illinois
Urbana, IL 61801
reply to: b-hannon@uiuc.edu
http://www.gis.uiuc.edu
217 333-0348 office
217 244 1785 fax