CMStatistics 2021: Start Registration
View Submission - CMStatistics
Title: Growth curves for multiple-output response variables via Bayesian quantile regression models Authors:  Bruno Santos - University of Kent (United Kingdom) [presenting]
Agatha Rodrigues - Federal University of Espirito Santo (Brazil)
Thomas Kneib - University of Goettingen (Germany)
Abstract: Reference fetal growth curves play an important role in identifying fetal growth restriction, macrosomia and other fetal malformations. This is verified based on percentiles of some biometric measurements at a specific gestational age using obstetric ultrasound. As an example, the diagnosis of microcephaly is based on a biparietal diameter smaller than the 10th percentile based on the reference curve. In practice, each biometric measurement reference curve is constructed independently of other measurements, even if they are correlated and some information about dependencies among them might be lost. Here we use these measurements to define growth curves modelling jointly more than one measurement. We consider structured additive quantile regression models for multiple-output response variables, where we are able to specify a nonlinear effect of time. We define a Markov Chain Monte Carlo (MCMC) procedure for model estimation, using ideas previously discussed in the literature. We examine four different ultrasound measurements and we show how one can retrieve more information when modelling these response variables jointly instead of individually. We illustrate the method with data from pregnancies from the University Hospital of the University of Sao Paulo (HU / USP) in the city of Sao Paulo, Brazil.