10.6084/M9.FIGSHARE.21641763
Yuefeng Han
Yuefeng
Han
Rong Chen
Rong
Chen
Cun-Hui Zhang
Cun-Hui
Zhang
Qiwei Yao
Qiwei
Yao
Simultaneous Decorrelation of Matrix Time Series
<p>We propose a contemporaneous bilinear transformation for a <i>p</i> × <i>q</i> matrix time series to alleviate the difficulties in modeling and forecasting matrix time series when <i>p</i> and/or <i>q</i> are large. The resulting transformed matrix assumes a block structure consisting of several small matrices, and those small matrix series are uncorrelated across all times. Hence, an overall parsimonious model is achieved by modeling each of those small matrix series separately without the loss of information on the linear dynamics. Such a parsimonious model often has better forecasting performance, even when the underlying true dynamics deviates from the assumed uncorrelated block structure after transformation. The uniform convergence rates of the estimated transformation are derived, which vindicate an important virtue of the proposed bilinear transformation, that is, it is technically equivalent to the decorrelation of a vector time series of dimension max(<i>p</i>, <i>q</i>) instead of <i>p</i> × <i>q</i>. The proposed method is illustrated numerically via both simulated and real data examples. <a href="https://doi.org/10.1080/01621459.2022.2151448" target="_blank">Supplementary materials</a> for this article are available online.</p>
Space Science
Medicine
Genetics
Biotechnology
Evolutionary Biology
Ecology
Mathematical Sciences not elsewhere classified
Inorganic Chemistry
Infectious Diseases
Taylor & Francis
2023
2023-01-11
2023-05-31
Dataset
760621 Bytes
10.1080/01621459.2022.2151448
CC BY 4.0