ABSTRACT
This chapter aims to give a sense of the breadth of applications of quantile regression in ecology and the environmental sciences. It examines in greater detail the models, inferences, and interpretations for applications. Initial interest in quantile regression in ecology was primarily motivated by the desire to model boundary-type relationships between variables to provide an interpretation consistent with the concept of limiting factors based on Liebig's law of the minimum. The chapter provides additional estimates for multiple sites within a watershed to examine differences occurring within a watershed. It demonstrates the use of a censored quantile regression estimator that provides a generalization of the Kaplan–Meier estimator to linear models. As the application on water quality trends demonstrated, the ability to use nonlinear transformations with linear quantile regression to model inherently nonlinear, multiplicative relationships without any of the bias-correcting adjustments needed for modeling means in this same framework is a great advantage for ecological and environmental modeling.
