Saved in:
Bibliographic Details
Main Authors: Li, Dong, Qiao, Xinghao, Wang, Zihan
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2311.02450
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908612952064000
author Li, Dong
Qiao, Xinghao
Wang, Zihan
author_facet Li, Dong
Qiao, Xinghao
Wang, Zihan
contents This paper addresses the fundamental task of estimating covariance matrix functions for high-dimensional functional data/functional time series. We consider two functional factor structures encompassing either functional factors with scalar loadings or scalar factors with functional loadings, and postulate functional sparsity on the covariance of idiosyncratic errors after taking out the common unobserved factors. To facilitate estimation, we rely on the spiked matrix model and its functional generalization, and derive some novel asymptotic identifiability results, based on which we develop DIGIT and FPOET estimators under two functional factor models, respectively. Both estimators involve performing associated eigenanalysis to estimate the covariance of common components, followed by adaptive functional thresholding applied to the residual covariance. We also develop functional information criteria for model selection with theoretical guarantees. The convergence rates of involved estimated quantities are respectively established for DIGIT and FPOET estimators. Numerical studies including extensive simulations and a real data application on functional portfolio allocation are conducted to examine the finite-sample performance of the proposed methodology.
format Preprint
id arxiv_https___arxiv_org_abs_2311_02450
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Factor-guided estimation of large covariance matrix function with conditional functional sparsity
Li, Dong
Qiao, Xinghao
Wang, Zihan
Methodology
This paper addresses the fundamental task of estimating covariance matrix functions for high-dimensional functional data/functional time series. We consider two functional factor structures encompassing either functional factors with scalar loadings or scalar factors with functional loadings, and postulate functional sparsity on the covariance of idiosyncratic errors after taking out the common unobserved factors. To facilitate estimation, we rely on the spiked matrix model and its functional generalization, and derive some novel asymptotic identifiability results, based on which we develop DIGIT and FPOET estimators under two functional factor models, respectively. Both estimators involve performing associated eigenanalysis to estimate the covariance of common components, followed by adaptive functional thresholding applied to the residual covariance. We also develop functional information criteria for model selection with theoretical guarantees. The convergence rates of involved estimated quantities are respectively established for DIGIT and FPOET estimators. Numerical studies including extensive simulations and a real data application on functional portfolio allocation are conducted to examine the finite-sample performance of the proposed methodology.
title Factor-guided estimation of large covariance matrix function with conditional functional sparsity
topic Methodology
url https://arxiv.org/abs/2311.02450