A gentle introduction to quantile regression for ecologists

By Brian S. Cade, B. Noon. 2003

1,870 citations recorded by the source corpus.

Subfields: Environmental Engineering, Nature and Landscape Conservation, Water Science and Technology

View Paper: https://www.semanticscholar.org/paper/8c0dc7f7362479553ac7747b66933a425b0840c8

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Environmental Engineering

In Nature and Landscape Conservation

In Water Science and Technology

Citations in the corpus

Complex hydraulic and substrate variables limit freshwater mussel species richness and abundance (2010, methodology): “Quantile regression estimates are semiparametric; no parametric distributional form is assumed for random errors but is assumed for the deterministic portion of the model (Cade and Noon 2003).”
Nitrogen uptake, fixation and response to fertilizer N in soybeans: A review (2008, methodology): “Quantile regression techniques were used to model empirical relationships among variables (Cade and Noon, 2003, Koenker and Hallock, 2001).”
Robustness and generalization (2010, background): “These loss functions are of interest, and have been applied in many areas including ecology (Cade and Noon 2003), medicine (Cole 1988) and finance (Koenker and Bassett 1978), because they are less sensitive to the presence of outliers than the standard average loss (Huber 1981).”
Yield Gap and Production Gap of Rainfed Winter Wheat in the Southern Great Plains (2014, methodology): “Similar approaches have been reported using the 95th percentile instead of the maximum value (Cade and Noon, 2003; Grassini et al., 2009).”
Range fidelity: The missing link between caribou decline and habitat alteration? (2010, background): “This result appears in opposition to conclusions obtained by Dalerum et al. (2007), who found no effect of wildfire on caribou home-range size and inter-annual range fidelity in Alberta.”
Interactions between water and nitrogen in Australian cropping systems: physiological, agronomic, economic, breeding and modelling perspectives (2016, methodology): “The data were further analysed by using percentile regression to capture the top and bottom boundaries of crop responses to fertiliser (Cade and Noon 2003).”
Comparing the Performance of Statistical Models for Predicting PM10 Concentrations (2014, methodology): “While Carslaw et al. (2007) and Westmoreland et al.”
Predicting spatial patterns of plant species richness: a comparison of direct macroecological and species stacking modelling approaches (2011, background): “The species number predicted by traditional MEM corresponds to the mean number of species that can actually occur at a place given macroecological conditions, but the maximum richness could also be predicted through quantile regression (Cade & Noon, 2003).”
Enforcement and over-compliance (2007, background): “We also find that average discharges for both BOD and TSS trend downward over time.”
Can species distribution modelling provide estimates of population densities? A case study with jaguars in the Neotropics (2012, methodology): “Because of the complex non-functional relationship between density and suitability, a quantile regression may produce better results than the OLS approach (for further information on quantile regression, see Cade & Noon, 2003).”
A partially collapsed Gibbs sampler for Bayesian quantile regression (2009, background): “The need for and success of quantile regression in ecology has been attributed to the complexity of interactions between difierent factors leading to data with unequal variation of one variable for difierent ranges of another variable ( Cade and Noon, 2003 ).”
A generalized class of skew distributions and associated robust quantile regression models (2014, methodology): “According to Koenker (2005, p.147), this quantile regression model is a quantile location-shift model of the form Qq(yi|xi) = x′iβ + σF−1(q), where Qq(yi|xi) is the conditional quantile or the inverse of the conditional cdf of yi given xi (Cade & Noon, 2003) and F−1(·) is an inverse cdf.”