Comments on: Shape-based functional data analysis III

• J. E. Borgert ^[1] ; J. S. Marron ^[1]
  1. [1] University of North Carolina at Chapel Hill

     University of North Carolina at Chapel Hill

     Township of Chapel Hill, Estados Unidos

     
• Localización: Test: An Official Journal of the Spanish Society of Statistics and Operations Research, ISSN-e 1863-8260, ISSN 1133-0686, Vol. 33, Nº. 1, 2024, págs. 66-70
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • This discussion paper applauds the authors for their impactful contribution to functional data analysis (FDA). Their primary insight lies in a formal mathematical definition of the “shape” of a curve, which they connect to familiar intuitive notions through a number of examples. Notably, the paper highlights the pitfalls of less well-thought-out curve registration approaches. The authors’ application of COVID-19 data enriches the discussion, highlighting the work’s practical relevance. We discuss connections of this work with object-oriented data analysis and propose enhancements to the authors’ shape-based functional principal component analysis. Additionally, we illustrate the practical significance of adaptive alignment with an example from our own research.