Differences
This shows you the differences between two versions of the page.
| Both sides previous revision Previous revision Next revision | Previous revision Next revision Both sides next revision | ||
|
group_modelling:goodnessoffit [2009/03/31 09:12] inpeifgi |
group_modelling:goodnessoffit [2009/06/09 14:45] inpeifgi |
||
|---|---|---|---|
| Line 18: | Line 18: | ||
| With these metrics, it is possible to calibrate or to validate the model. | With these metrics, it is possible to calibrate or to validate the model. | ||
| In fact, the final objective of these goodness-of-fit methods is to point out how to improve the model. | In fact, the final objective of these goodness-of-fit methods is to point out how to improve the model. | ||
| + | |||
| =====State of the Art===== | =====State of the Art===== | ||
| Line 30: | Line 31: | ||
| Scale (here in terms of resolution and extend) is an important property of LUCC models. Li (2000) investigated the fractal properties which are typical to many land use (change) patterns. Another approach by Jantz and Goetz (2005) compared different goodness-of-fit measures on several resolutions for an urban growth model as land use changes may show varying behaviour on different scales. But also extend can change models a lot as Kok and Veldkamp (2001) showed for national vs. multinational models. | Scale (here in terms of resolution and extend) is an important property of LUCC models. Li (2000) investigated the fractal properties which are typical to many land use (change) patterns. Another approach by Jantz and Goetz (2005) compared different goodness-of-fit measures on several resolutions for an urban growth model as land use changes may show varying behaviour on different scales. But also extend can change models a lot as Kok and Veldkamp (2001) showed for national vs. multinational models. | ||
| - | Calibration of cellular automata or agent-based models is not a trivial task as parameters influence is in most cases non-linear and often the number of parameters is high, making comprehensive evaluation of all combinations unfeasible. Simple approaches like by Clarke et al. (1998) generate lots of simulations to be evaluated by the user, they consider interactive visualization as an important tool. Multi-resolution search of the parameter space as described in Candau (2002) may help to detect important parameter combinations and subsequently to adjust them with feasible computational effort. Still users may not find the most influential parameter combinations. This task was addressed by Miller (1998) who used several robust optimization algorithms to investigate the parameter space. Wu (2002) uses the data to fit a prior distribution to the parameters and updates it according to the results of Monte Carlo simulations for calibrating a cellular automata model. | + | Calibration of cellular automata or agent-based models is not a trivial task as parameters influence is in most cases non-linear and often the number of parameters is high, making comprehensive evaluation of all combinations unfeasible. Simple approaches like by Clarke et al. (1998) generate lots of simulations to be evaluated by the user, they consider interactive visualization as an important tool. Multi-resolution search of the parameter space as described in Candau (2002) or Hakan et.al (2007) may help to detect important parameter combinations and subsequently to adjust them with feasible computational effort. Still users may not find the most influential parameter combinations. This task was addressed by Miller (1998) who used several robust optimization algorithms to investigate the parameter space. Wu (2002) uses the data to fit a prior distribution to the parameters and updates it according to the results of Monte Carlo simulations for calibrating a cellular automata model. Whereas calibration of agent based models is still a domain of econometrics (e.g. Rogers & von Tessin 2004). |
| The diversity of LUCC models may require different calibration and validation methods. An overview over current models is given by Agarwal et al., a comparison of several models by Pontius et al. (2008). Parker et al. (2003) focus on multi-agent models only. | The diversity of LUCC models may require different calibration and validation methods. An overview over current models is given by Agarwal et al., a comparison of several models by Pontius et al. (2008). Parker et al. (2003) focus on multi-agent models only. | ||
| Line 134: | Line 135: | ||
| pp. 795–818. | pp. 795–818. | ||
| - | ==not directly relevant== | + | ==related literature== |
| GILES, R. H., Jr. & TRANI, M. K..1999. {{http://www.springerlink.com/content/je6g2v0khpl93gpc/fulltext.pdf|Key elements of landscape pattern measures}}. Environmental Management 23 (4):477–81.) | GILES, R. H., Jr. & TRANI, M. K..1999. {{http://www.springerlink.com/content/je6g2v0khpl93gpc/fulltext.pdf|Key elements of landscape pattern measures}}. Environmental Management 23 (4):477–81.) | ||
