B1482
Title: Online detection of changes in moment-based projections: When to retrain deep learners or update portfolios
Authors: Ansgar Steland - RWTH Aachen University (Germany) [presenting]
Abstract: Sequential monitoring of high-dimensional nonlinear time series is studied for a projection of the second-moment matrix, a problem interesting in its own right and specifically arising in deep learning and finance. Open-end, as well as closed-end monitoring, is studied under mild assumptions on the training sample and the observations of the monitoring period. Asymptotics is based on Gaussian approximations of projected partial sums, allowing for an estimated projection vector. Estimation is studied both for classical non-sparsity as well as under sparsity. For the case that the optimal projection depends on the unknown covariance matrix, hard- and soft-thresholded estimators are studied. Applications in finance and training of deep neural networks are discussed. The proposed detectors typically reduce the required computational costs, as illustrated by monitoring synthetic data.
