Calibrating Expert Assessments Using Hierarchical Gaussian Process Models
Expert assessments are routinely used to inform management and other decision making. However, often these assessments contain considerable biases and uncertainties for which reason they should be calibrated if possible. Moreover, coherently combining multiple expert assessments into one estimate poses a long-standing problem in statistics since modeling expert knowledge is often difficult. Here, we present a hierarchical Bayesian model for expert calibration in a task of estimating a continuous univariate parameter. The model allows experts' biases to vary as a function of the true value of the parameter and according to the expert's background. We follow the fully Bayesian approach (the s…
Eliciting expert knowledge to inform stock status for data-limited stock assessments
Data-limited fisheries are a major challenge for stock assessment analysts, as many traditional data-rich models cannot be implemented. Approaches based on stock reduction analysis offer simple ways to handle low data availability, but are particularly sensitive to assumptions on relative stock status (i.e., current biomass compared to unperturbed biomass). For the vast majority of data-limited stocks, stock status is unmeasured. The present study presents a method to elicit expert knowledge to inform stock status and a novel, user-friendly on-line application for expert elicitation. Expert opinions are compared to stock status derived from data-rich models. Here, it is evaluated how expert…