Regularized Regression: A New Tool for Investigating and Predicting Tree Growth

Neighborhood models have allowed us to test many hypotheses regarding the drivers of variation in tree growth, but require considerable computation due empirically supported non-linear relationships they include. Regularized regression represents a far more efficient neighborhood modeling method, it is unclear whether such an ecologically unrealistic model can provide accurate insights on growth. Rapid becoming increasingly important as ecological datasets grow size, and may be essential when using predict growth beyond sample plots or into future. We built novel regularized investigated reached same conclusions commonly used model, how influenced by species identity neighboring trees. also evaluated ability both interpolate trees not included fitting dataset. Our replicated most classical model’s inferences fraction time without high-performance computing resources. found that methods could out-of-sample method making predictions varied among focal species. particularly for comparing because automates process selection handle correlated explanatory variables. This feature means select potential variables (e.g., climate variables) thereby streamline development model. Both future research must determine extrapolated experiencing conditions. Overall, we conclude complement investigation represent valuable tool advancing this field toward prediction.