Version 3 2021-12-16, 13:40Version 3 2021-12-16, 13:40
Version 2 2020-10-12, 12:50Version 2 2020-10-12, 12:50
Version 1 2020-09-08, 13:00Version 1 2020-09-08, 13:00
journal contribution
posted on 2021-12-16, 13:40authored byHolger Dette, Weichi Wu
We develop an estimator for the high-dimensional covariance matrix of a locally stationary process with a smoothly varying trend and use this statistic to derive consistent predictors in nonstationary time series. In contrast to the currently available methods for this problem the predictor developed here does not rely on fitting an autoregressive model and does not require a vanishing trend. The finite sample properties of the new methodology are illustrated by means of a simulation study and a financial indices study. Supplementary materials for this article are available online.
Funding
Holger Dette gratefully acknowledges Collaborative Research Center “Statistical modeling of nonlinear dynamic processes” (SFB 823, Project A1, C1) of the German Research Foundation (DFG). Weichi Wu gratefully acknowledges NSFC Young Program (no. 11901337) and BJNSF (no. Z190001).