Results 1 
4 of
4
Bayesian Calibration of Computer Models
 Journal of the Royal Statistical Society, Series B, Methodological
, 2000
"... this paper a Bayesian approach to the calibration of computer models. We represent the unknown inputs as a parameter vector `. Using the observed data we derive the posterior distribution of `, which in particular quantifies the `residual uncertainty' about ..."
Abstract

Cited by 192 (3 self)
 Add to MetaCart
this paper a Bayesian approach to the calibration of computer models. We represent the unknown inputs as a parameter vector `. Using the observed data we derive the posterior distribution of `, which in particular quantifies the `residual uncertainty' about
Bayesian uncertainty assessment in multicompartment deterministic simulation models for environmental risk assessment
, 2003
"... We use a special case of Bayesian melding to make inference from deterministic models while accounting for uncertainty in the inputs to the model. The method uses all available information, based on both data and expert knowledge, and extends current methods of ‘uncertainty analysis ’ by updating mo ..."
Abstract

Cited by 8 (0 self)
 Add to MetaCart
We use a special case of Bayesian melding to make inference from deterministic models while accounting for uncertainty in the inputs to the model. The method uses all available information, based on both data and expert knowledge, and extends current methods of ‘uncertainty analysis ’ by updating models using available data. We extend the methodology for use with sequential multicompartment models. We present an application of these methods to deterministic models for concentration of polychlorinated biphenyl (PCB) in soil and vegetables. The results are posterior distributions of concentration in soil and vegetables which account for all available evidence and uncertainty. Model uncertainty is not considered. Copyright # 2003 John Wiley & Sons, Ltd. key words: bayesian melding; deterministic models; sampling importance resampling; uncertainty analysis 1. BACKGROUND Risk is the probability of an adverse outcome, and risk assessment is a process by which an estimate of this probability is obtained. A health risk assessment generally follows four steps: hazard identification, dose–response assessment, exposure assessment and risk characterization. In the hazard identification step, the potential hazard and its adverse health effects (if any) are identified. The exposure assessment step identifies populations which could be exposed and the pathways by which
Model uncertainty;
"... recharge models developed for the Death Valley regional flow system (DVRFS), covering is uncertain which recharge model (or models) should be used as input for groundwater betweenexpert variability. The most favorable model, on average, is the most compliond highest prior probability. The aggregat ..."
Abstract
 Add to MetaCart
(Show Context)
recharge models developed for the Death Valley regional flow system (DVRFS), covering is uncertain which recharge model (or models) should be used as input for groundwater betweenexpert variability. The most favorable model, on average, is the most compliond highest prior probability. The aggregated prior probabilities are close to the neutral and the largest amount of information used to evaluate the models. However, when enough data are available, we prefer to use a crossvalidation method to select the best set of prior model probabilities that gives the best predictive performance.