Font Size: 

Quantile regression allows estimation of the relationship between response and explanatory variables at any percentile of the distribution of the response (conditioned on the explanatory variables). We extend quantile regression to the functional case, rewriting the quantile regression model as a generalized additive model where both the functional covariates and the functional coefficients are parametrized in terms of B-splines. Parameter estimation is done using a penalized iterative reweighed least squares (PIRLS) algorithm. We evaluate the performance of the model by means of a simulation study.