FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2216977548> ?p ?o ?g. }

• W2216977548 startingPage "3" @default.
• W2216977548 abstract "When applying a hydrological model to a catchment it can be difficult to choose from the variety of models that exist. This difficulty can be amplified when considering the uncertainty surrounding the parameters of the model. Few studies have compared the performance of hydrological models applied to a common catchment, nor have they provided a framework within which modelers may choose and implement a model based on its performance for a specific catchment. An alternative approach to selecting a single model is to combine the results from several hydrological models. Bayesian methods can provide an ideal means to compare and combine competing models whilst allowing for model uncertainty. In comparing two models, the traditional approach requires calculation of the Bayes Factor, which is the posterior probability ratio of the models (assuming equal prior probabilities). Calculation of Bayes factors is complicated by the computational effort required, particularly for high dimensional models. Simpler methods of combining models include a simple or weighted average of the results, or via an extension of calculating Bayes factors. Bayesian model averaging involves combination of individual models with weights proportional to the posterior probability of the model. A method that allows for more complex combination of results than that provided by Bayesian model averaging is presented in hierarchical mixtures of experts (HME) models. HME models provide an improvement of simple combinations of models, by allowing the way that model predictions are combined to depend on predictor variables. Models are then combined depending on catchment conditions as summarized by the predictor variables. This study develops a catchment specific model aggregation framework for hydrological models. The study builds on previous work in which the parameters of the abc model were estimated using computations carried out via Markov chain Monte Carlo methods. The approach used daily rainfall, runoff and evapotranspiration data from the Bass River watershed, located in the South Gippsland Basin in Victoria to determine which formulation best represents the catchment. The framework uses an adaptive Metropolis algorithm to calculate the models posterior odds to enable direct model comparison. The study introduces the HME architecture, and compares results using different predictor variables." @default.
• W2216977548 created "2016-06-24" @default.
• W2216977548 creator A5005139027 @default.
• W2216977548 creator A5011403405 @default.
• W2216977548 creator A5038287899 @default.
• W2216977548 date "2003-01-01" @default.
• W2216977548 modified "2023-09-26" @default.
• W2216977548 title "Bayesian Inference in a Hydrological Application of Hierarchical Mixtures of Experts" @default.
• W2216977548 hasPublicationYear "2003" @default.
• W2216977548 type Work @default.
• W2216977548 sameAs 2216977548 @default.
• W2216977548 citedByCount "0" @default.
• W2216977548 crossrefType "journal-article" @default.
• W2216977548 hasAuthorship W2216977548A5005139027 @default.
• W2216977548 hasAuthorship W2216977548A5011403405 @default.
• W2216977548 hasAuthorship W2216977548A5038287899 @default.
• W2216977548 hasConcept C107673813 @default.
• W2216977548 hasConcept C119857082 @default.
• W2216977548 hasConcept C124101348 @default.
• W2216977548 hasConcept C142291917 @default.
• W2216977548 hasConcept C144986985 @default.
• W2216977548 hasConcept C154945302 @default.
• W2216977548 hasConcept C160234255 @default.
• W2216977548 hasConcept C191413810 @default.
• W2216977548 hasConcept C207201462 @default.
• W2216977548 hasConcept C2776214188 @default.
• W2216977548 hasConcept C41008148 @default.
• W2216977548 hasConcept C57830394 @default.
• W2216977548 hasConceptScore W2216977548C107673813 @default.
• W2216977548 hasConceptScore W2216977548C119857082 @default.
• W2216977548 hasConceptScore W2216977548C124101348 @default.
• W2216977548 hasConceptScore W2216977548C142291917 @default.
• W2216977548 hasConceptScore W2216977548C144986985 @default.
• W2216977548 hasConceptScore W2216977548C154945302 @default.
• W2216977548 hasConceptScore W2216977548C160234255 @default.
• W2216977548 hasConceptScore W2216977548C191413810 @default.
• W2216977548 hasConceptScore W2216977548C207201462 @default.
• W2216977548 hasConceptScore W2216977548C2776214188 @default.
• W2216977548 hasConceptScore W2216977548C41008148 @default.
• W2216977548 hasConceptScore W2216977548C57830394 @default.
• W2216977548 hasLocation W22169775481 @default.
• W2216977548 hasOpenAccess W2216977548 @default.
• W2216977548 hasPrimaryLocation W22169775481 @default.
• W2216977548 hasRelatedWork W1518085804 @default.
• W2216977548 hasRelatedWork W1679242182 @default.
• W2216977548 hasRelatedWork W1761851204 @default.
• W2216977548 hasRelatedWork W177094713 @default.
• W2216977548 hasRelatedWork W1952472251 @default.
• W2216977548 hasRelatedWork W1984138429 @default.
• W2216977548 hasRelatedWork W2004883664 @default.
• W2216977548 hasRelatedWork W2071405969 @default.
• W2216977548 hasRelatedWork W2072634742 @default.
• W2216977548 hasRelatedWork W2117670920 @default.
• W2216977548 hasRelatedWork W2142245349 @default.
• W2216977548 hasRelatedWork W2142443481 @default.
• W2216977548 hasRelatedWork W2159293413 @default.
• W2216977548 hasRelatedWork W2228337672 @default.
• W2216977548 hasRelatedWork W2316408149 @default.
• W2216977548 hasRelatedWork W2608757158 @default.
• W2216977548 hasRelatedWork W2797733738 @default.
• W2216977548 hasRelatedWork W72477001 @default.
• W2216977548 hasRelatedWork W2049570168 @default.
• W2216977548 hasRelatedWork W2481109466 @default.
• W2216977548 isParatext "false" @default.
• W2216977548 isRetracted "false" @default.
• W2216977548 magId "2216977548" @default.
• W2216977548 workType "article" @default.