On change point detection using the fused lasso method
This work addresses change point detection, a critical problem in fields like science and engineering, but it is incremental as it builds on existing fused lasso methods by analyzing specific conditions for consistency.
The paper analyzes the asymptotic properties of l1 penalized maximum likelihood estimation for detecting change points in non-stationary time series with piece-wise constant means and/or variances, establishing that the fused lasso signal approximator achieves sparse consistency only when consecutive changes have alternating signs, with rates of convergence provided.
In this paper we analyze the asymptotic properties of l1 penalized maximum likelihood estimation of signals with piece-wise constant mean values and/or variances. The focus is on segmentation of a non-stationary time series with respect to changes in these model parameters. This change point detection and estimation problem is also referred to as total variation denoising or l1 -mean filtering and has many important applications in most fields of science and engineering. We establish the (approximate) sparse consistency properties, including rate of convergence, of the so-called fused lasso signal approximator (FLSA). We show that this only holds if the sign of the corresponding consecutive changes are all different, and that this estimator is otherwise incapable of correctly detecting the underlying sparsity pattern. The key idea is to notice that the optimality conditions for this problem can be analyzed using techniques related to brownian bridge theory.