Boosting the Scalability of Farm-Level Models: Efficient Surrogate Modeling of Compositional Simulation Output

Abstract Surrogate modeling can overcome computational and data-privacy constraints of micro-scale economic models support their incorporation into large-scale simulations interactive simulation experiments. We compare four data-driven methods to reproduce the aggregated crop area response simulated by farm-level in price variation. use isometric log-ratio transformation accommodate compositional nature output sequential sampling with stability analysis for efficient model selection. Extreme gradient boosting outperforms multivariate adaptive regressions splines, random forest regression, classical multinomial-logistic regression achieves high goodness-of-fit from moderately sized samples. Explicitly including ratio terms between input variables considerably improved prediction, even highly automatic machine learning that should principle be able detect such variable interaction automatically. The presented methodology provides a solid basis surrogate integrated experiments stakeholders.