Title: a Working Guide to Boosted Regression Trees

Comment: This is the final submitted manuscript for this paper, without further corrections. It has been reformatted for efficient printing. For a pdf of the final Blackwell publishing version please email Jane Elith SUMMARY 1. Ecologists use statistical models for both explanation and prediction, and need techniques that are flexible enough to express typical features of their data such as non-linearities and interactions. 2. This study provides a working guide to boosted regression trees (BRT), an ensemble method for fitting statistical models that differs fundamentally from conventional techniques that aim to fit a single parsimonious model. BRT combines the strengths of two algorithms: regression trees (models that relate a response to its predictors by recursive binary splits) and boosting (an adaptive method for combining many simple models to give improved predictive performance). The final BRT model can be understood as an additive regression model in which individual terms are simple trees, fitted in a forward, stage-wise fashion. 3. BRT incorporates important advantages of tree-based methods, handling different types of predictor variables and accommodating missing data. It has no need for prior data transformation or elimination of outliers, can fit complex non-linear relationships, and automatically handles interaction effects between predictors. Fitting multiple trees in BRT overcomes the biggest drawback of single tree models, their relatively poor predictive performance. Even though BRT models are complex, they can be summarised in ways that give powerful ecological insight, and their predictive performance is superior to most traditional modelling methods. 4. The unique features of BRT raise a number of practical issues in model fitting. We demonstrate the practicalities and advantages of using BRT through a distributional analysis of the short-finned eel (Anguilla australis), a native freshwater fish of New Zealand. We use a dataset of over 13,000 sites to illustrate effects of several settings, and then fit and interpret a model using a subset of the data. We provide code and a tutorial to enable the wider use of BRT by ecologists.