A0477
Title: Order determination in second-order source separation models using data augmentation
Authors: Una Radojicic - Technical University of Vienna (Austria)
Klaus Nordhausen - University of Helsinki (Finland) [presenting]
Abstract: A robust estimator is proposed for determining the number of latent components in an internal noise model within the second-order source separation (SOS) framework. The method incorporates a data augmentation strategy along with the robust SOS approach, eSAM-AMUSE, which utilizes information from eigenvalues and variations of eigenvectors. The dimension estimate derived from the approach can be visualized using a ladle plot. Through a simulation study, the new estimator is demonstrated to exhibit superior properties and consistently outperform the bootstrap-based AMUSEladle estimator.
