Title: Quantile regression for longitudinal functional data with application to feed intake of lactating sows
Authors: Maria Laura Battagliola - EPFL (Switzerland) [presenting]
Helle Sorensen - University of Copenhagen (Denmark)
Anders Tolver - University of Copenhagen (Denmark)
Ana-Maria Staicu - North Carolina State University (United States)
Abstract: A model framework and estimation methodology are introduced for quantile regression in scenarios with clustered or longitudinal data and functional covariates. The proposed quantile regression model uses a time-varying regression coefficient function to quantify the association between covariates and quantile level of interest, and includes subject-specific intercepts to incorporate within-subject dependence. Estimation relies on spline representation of the unknown coefficient functions, and can be carried out with existing software. The proposed method is studied numerically in a simulation study that covers a wide range of situations, and bootstrap procedures for bias adjustment and computation of standard errors are introduced. The work is motivated by a study on lactating sows, where the main interest is the influence of temperature, measured throughout the day, on the lower quantiles of feed intake. Analysis of the lactation data indicates, among others, that the influence of temperature increases during the lactation period.