WEBINAR: Predicting willingness-to-pay values for mammals and birds using machine-learning
DECIPHER project recently hosted an insightful webinar on the application of machine learning to predict willingness-to-pay (WTP) values for the conservation of bird and mammal species. Led by Maarten Broekman from Radboud University the session explored innovative methods for incorporating biodiversity value into economic decision-making.
Biodiversity loss remains a critical global challenge, with non-market values—such as the intrinsic worth of species or their ecological importance—often overlooked in traditional economic models. Addressing this gap, the webinar introduced the use of eXtreme Gradient Boosting (XGBoost), a machine-learning algorithm, and the integration species-specific characteristics in this model to improve WTP predictions.
During the session, Broekman highlighted the comprehensive data collection process, which involved analyzing over 2,000 studies and extracting WTP values for 81 bird and mammal species. The model incorporated various predictors, including body mass, popularity, image counts, threat status, and ecological factors, ultimately demonstrating superior predictive accuracy compared to conventional linear models.
Key takeaways from the discussion emphasized the significance of addressing uncertainties in WTP estimates, and the potential integration of these findings into economic frameworks such as the DECIPHER project’s biodiversity assessment tools.
This discussion is a crucial step towards bridging ecological and economic considerations in policy-making.
Stay tuned for more DECIPHER events by subscribing to the project Newsletter.