Functional data analysis with increasing number of projections
From MaRDI portal
Abstract: Functional principal components (FPC's) provide the most important and most extensively used tool for dimension reduction and inference for functional data. The selection of the number, d, of the FPC's to be used in a specific procedure has attracted a fair amount of attention, and a number of reasonably effective approaches exist. Intuitively, they assume that the functional data can be sufficiently well approximated by a projection onto a finite-dimensional subspace, and the error resulting from such an approximation does not impact the conclusions. This has been shown to be a very effective approach, but it is desirable to understand the behavior of many inferential procedures by considering the projections on subspaces spanned by an increasing number of the FPC's. Such an approach reflects more fully the infinite-dimensional nature of functional data, and allows to derive procedures which are fairly insensitive to the selection of d. This is accomplished by considering limits as d tends to infinity with the sample size. We propose a specific framework in which we let d tend to infinity by deriving a normal approximation for the two-parameter partial sum process of the scores xi_{i,j} of the i-th function with respect to the j-th FPC. Our approximation can be used to derive statistics that use segments of observations and segments of the FPC's. We apply our general results to derive two inferential procedures for the mean function: a change-point test and a two-sample test. In addition to the asymptotic theory, the tests are assessed through a small simulation study and a data example.
Recommendations
- A two-sample test for mean functions with increasing number of projections
- Detecting Changes in the Mean of Functional Observations
- On Properties of Functional Principal Components Analysis
- An introduction to functional data analysis and a principal component approach for testing the equality of mean curves
- Selecting the number of principal components in functional data
Cites work
- scientific article; zbMATH DE number 3858075 (Why is no real title available?)
- scientific article; zbMATH DE number 3725366 (Why is no real title available?)
- scientific article; zbMATH DE number 741240 (Why is no real title available?)
- scientific article; zbMATH DE number 1048663 (Why is no real title available?)
- scientific article; zbMATH DE number 3428313 (Why is no real title available?)
- Assessing the Finite Dimensionality of Functional Data
- Break detection in the covariance structure of multivariate time series models
- Central limit theorems in D[0, 1]
- Classes of linear operators. Vol. I
- Common functional principal components
- Detecting Changes in the Mean of Functional Observations
- Estimation and testing for spatially indexed curves with application to ionospheric and magnetic field trends
- Estimation of the Mean of Functional Time Series and a Two-Sample Problem
- Evaluating stationarity via change-point alternatives with applications to fMRI data
- Extensions of results of Komlós, Major, and Tusnády to the multivariate case
- Functional Data Analysis with R and MATLAB
- Functional data analysis.
- Identifying the finite dimensionality of curve time series
- Inference for functional data with applications
- Normal Approximation
- Parametric statistical change point analysis. With applications to genetics, medicine, and finance
- Second-order comparison of Gaussian random functions and the geometry of DNA minicircles
- Strong invariance principles for partial sums of independent random vectors
- Testing Hypotheses in the Functional Linear Model
- The Existence of Probability Measures with Given Marginals
- The invariance principle for Banach space valued random variables
- Two sample inference in functional linear models
Cited in
(24)- Asymptotic normality of a generalized maximum mean discrepancy estimator
- Robust multivariate functional discriminant coordinates
- Two-sample tests for multivariate functional data with applications
- Inferential procedures for partially observed functional data
- Assessing the Finite Dimensionality of Functional Data
- Robust estimation in canonical correlation analysis for multivariate functional data
- An introduction to functional data analysis and a principal component approach for testing the equality of mean curves
- Kernel based method for the k-sample problem with functional data
- White noise testing and model diagnostic checking for functional time series
- Quadratic forms of the empirical processes for the two-sample problem for functional data
- Functional data analysis in the Banach space of continuous functions
- A Darling-Erdős-type CUSUM-procedure for functional data
- A randomness test for functional panels
- Detecting structural breaks in eigensystems of functional time series
- Linear Hypothesis Testing With Functional Data
- Optimal eigen expansions and uniform bounds
- Robust change-point detection for functional time series based on \(U\)-statistics and dependent wild bootstrap
- Testing for stationarity of functional time series in the frequency domain
- A \(k\)-sample test for functional data based on generalized maximum mean discrepancy
- Selecting the number of principal components in functional data
- A two-sample test for mean functions with increasing number of projections
- Functional Principal Component Regression and Functional Partial Least‐squares Regression: An Overview and a Comparative Study
- Multivariate analysis of variance for functional data
- Thresholding mean test for functional data with power enhancement
This page was built for publication: Functional data analysis with increasing number of projections
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q392095)
