[Back]

U. Radojicic, N. Lietzen, K. Nordhausen, J. Virta:
"Order Determination for Matrix-valued Observations Using Data Augmentation";
Talk: Statistical Seminar, J. J. Strossmayer University of Osijek Department of Mathematics, Osijek, Croatia; 2021-10-27.

We propose an automated way of determining the optimal number of low-rank components in dimension reduction of image data. The method is based on the combination of two-dimensional principal component analysis and an augmentation estimator proposed recently in the literature. Intuitively, the main idea is to combine a scree plot with information extracted from the eigenvectors of a variation matrix. Simulation studies show that the method provides accurate estimates and a demonstration with a finger data set showcases its performance in practice.

Dimensionality reduction , Fingers , Estimation , Signal processing , Data models , Data mining , Principal component analysis