Expert elicitation of autocorrelated time series with application to e3 (energy-environment-economic) forecasting models. (February 2017)

• Record Type: Journal Article
• Authors: Durbach, Ian; Merven, Bruno; McCall, Bryce
• Is Part Of: Environmental modelling & software. Volume 88(2017)
• Page Start: 93
• Page End: 105
• Publication Date: 2017-02
• Subjects: Forecasting -- Expert elicitation -- Greenhouse gas emissions -- e3 modelling -- Uncertainty; Environmental monitoring -- Computer programs -- Periodicals; Ecology -- Computer simulation -- Periodicals; Digital computer simulation -- Periodicals; Computer software -- Periodicals; Environmental Monitoring -- Periodicals; Computer Simulation -- Periodicals; Environnement -- Surveillance -- Logiciels -- Périodiques; Écologie -- Simulation, Méthodes de -- Périodiques; Simulation par ordinateur -- Périodiques; Logiciels -- Périodiques; Computer software; Digital computer simulation; Ecology -- Computer simulation; Environmental monitoring -- Computer programs; Periodicals; Electronic journals; 363.70015118
• DOI: 10.1016/j.envsoft.2016.11.007
• Languages: English
• ISSNs: 1364-8152

Abstract

Abstract: Explicitly representing uncertainty is recognised as a fundamental requirement of any long-term forecast. We propose and illustrate an expert elicitation protocol for constructing long-term probabilistic projections. Each projection represents a possible realization of a time series with autocorrelation properties, and thus a plausible future evolution of a quantity of interest. We illustrate the approach using two quantities – GDP growth rates and coal prices – that were elicited as part of a project producing baseline forecasts of greenhouse gas emissions in South Africa to 2050. The elicited projections can be used as inputs to deterministic structural models of the energy, economic, and environmental sectors (e3 or energy-environment-economic models), to generate similar probabilistic projections for any desired outputs of the e3 model. An R package for the generation and visualization of these probabilistic projections is provided. Highlights: Projections are time series that describe how a variable might evolve over time. We describe an expert elicitation protocol for assessing uncertainty around projections. The resulting probabilistic projections can be passed to e3 models to produce probabilistic projections of model outputs. The protocol is illustrated with two real-world case studies. An R package projections for assisting with the elicitation protocol is provided.