A0657
Title: Fussed nearly-isotonic signal approximation
Authors: Vladimir Pastukhov - Chalmers University of Technology (Sweden) [presenting]
Abstract: The aim is to estimate sparse piecewise monotone signals and introduce fussed nearly-isotonic signal approximation. We provide different numerical solutions to the optimisation problem, show that it is computationally feasible, derive the degrees of freedom for the estimator and study the asymptotic properties. Based on the simulations we show that the estimator behaves well and in certain cases, it outperforms other constrained estimators.
