Estimation for Latent Factor Models for High-Dimensional Time Series
Abstract
This paper deals with the dimension reduction for high-dimensional time series based on common factors. In particular we allow the dimension of time series to be as large as, or even larger than, the sample size . The estimation for the factor loading matrix and the factor process itself is carried out via an eigenanalysis for a non-negative definite matrix. We show that when all the factors are strong in the sense that the norm of each column in the factor loading matrix is of the order , the estimator for the factor loading matrix, as well as the resulting estimator for the precision matrix of the original -variant time series, are weakly consistent in -norm with the convergence rates independent of . This result exhibits clearly that the `curse' is canceled out by the `blessings' in dimensionality. We also establish the asymptotic properties of the estimation when not all factors are strong. For the latter case, a two-step estimation procedure is preferred accordingly to the asymptotic theory. The proposed methods together with their asymptotic properties are further illustrated in a simulation study. An application to a real data set is also reported.
Cite
@article{arxiv.1004.2138,
title = {Estimation for Latent Factor Models for High-Dimensional Time Series},
author = {Clifford Lam and Qiwei Yao and Neil Bathia},
journal= {arXiv preprint arXiv:1004.2138},
year = {2010}
}
Comments
35 pages article, 4 figures