Preprint
Abstract
Nowadays many real-world datasets can be considered as functional, in the sense that the processes which generate them are continuous. A fundamental property of this type of data is that in theory they belong to an infinite-dimensional space. Although in practice we usually receive finite observations, they are still high-dimensional and hence dimensionality reduction methods are crucial. In this vein, the main state-of-the-art method for functional data analysis is Functional PCA. Nevertheless, this classic technique assumes that the data lie in a linear manifold, and hence it could have problems when this hypothesis is not fulfilled. In this research, attention has been placed on a non-linear manifold learning method: Diffusion Maps. The article explains how to extend this multivariate method to functional data and compares its behavior against Functional PCA over different simulated and real examples.Keywords
All Related Versions
This publication has 9 references indexed in Scilit:
- Nonlinear Functional Modeling Using Neural NetworksJournal of Computational and Graphical Statistics, 2023
- Time coupled diffusion mapsApplied and Computational Harmonic Analysis, 2018
- Multivariate time-series analysis and diffusion mapsSignal Processing, 2015
- Audio-Visual Voice Activity Detection Using Diffusion MapsIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2015
- Applications of functional data analysis: A systematic reviewBMC Medical Research Methodology, 2013
- Nonlinear manifold representations for functional dataThe Annals of Statistics, 2012
- Diffusion mapsApplied and Computational Harmonic Analysis, 2006
- Support vector machine for functional data classificationNeurocomputing, 2006
- A Global Geometric Framework for Nonlinear Dimensionality ReductionScience, 2000