Monitoring for a Phase Transition in a Time Series of Wigner Matrices
By: Nina Dörnemann , Piotr Kokoszka , Tim Kutta and more
Potential Business Impact:
Find hidden changes in data streams.
We develop methodology and theory for the detection of a phase transition in a time-series of high-dimensional random matrices. In the model we study, at each time point \( t = 1,2,\ldots \), we observe a deformed Wigner matrix \( \mathbf{M}_t \), where the unobservable deformation represents a latent signal. This signal is detectable only in the supercritical regime, and our objective is to detect the transition to this regime in real time, as new matrix--valued observations arrive. Our approach is based on a partial sum process of extremal eigenvalues of $\mathbf{M}_t$, and its theoretical analysis combines state-of-the-art tools from random-matrix-theory and Gaussian approximations. The resulting detector is self-normalized, which ensures appropriate scaling for convergence and a pivotal limit, without any additional parameter estimation. Simulations show excellent performance for varying dimensions. Applications to pollution monitoring and social interactions in primates illustrate the usefulness of our approach.
Similar Papers
Deformation of semi-circle law for the correlated time series and Phase transition
Statistical Mechanics
Finds patterns in messy money data.
The Algorithmic Phase Transition in Symmetric Correlated Spiked Wigner Model
Statistics Theory
Find hidden patterns in noisy data using two sources.
The Algorithmic Phase Transition in Symmetric Correlated Spiked Wigner Model
Statistics Theory
Find hidden patterns in noisy data using two sources.