Abstract
Traditionally, uncertainty analysis of complex simulation models has been conducted based on the assumption of that the components of the model are independent. In practice, correlation is universal in ecosystems. This study applied Bayesian estimation and rejection sampling to generate correlated random samples for an uncertainty analysis of a process based forest growth model, a pipe model. Comparison of error budgets built using independent and correlated distributions shows that correlated distributions are very important to provide reasonable and realistic simulation and uncertainty analysis.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Recommended Citation
Fang, Shoufan and Gertner, George Z.
(2000).
"UNCERTAINTY ANALYSIS OF A PIPE MODEL BASED ON CORRELATED DISTRIBUTIONS,"
Conference on Applied Statistics in Agriculture.
https://doi.org/10.4148/2475-7772.1250
UNCERTAINTY ANALYSIS OF A PIPE MODEL BASED ON CORRELATED DISTRIBUTIONS
Traditionally, uncertainty analysis of complex simulation models has been conducted based on the assumption of that the components of the model are independent. In practice, correlation is universal in ecosystems. This study applied Bayesian estimation and rejection sampling to generate correlated random samples for an uncertainty analysis of a process based forest growth model, a pipe model. Comparison of error budgets built using independent and correlated distributions shows that correlated distributions are very important to provide reasonable and realistic simulation and uncertainty analysis.
