Fast Bayesian FPCA published in JCGS
papers
Joe Sartini, Xinkai Zhou, Elizabeth Selvin, Scott Zeger, and Ciprian Crainiceanu published Fast Bayesian Functional Principal Components Analysis in the Journal of Computational and Graphical Statistics. FAST is a fully Bayesian FPCA that accounts for uncertainty in the estimated principal components, motivated by mealtime glucose variability in the DASH4D continuous glucose monitoring study. Code is available as supplementary material.