A Variational View on Statistical Multiscale Estimation

Abstract

We present a unifying view on various statistical estimation techniques including penalization, variational and thresholding methods. These estimators will be analyzed in the context of statistical linear inverse problems including nonparametric and change point regression, and high dimensional linear models as examples. Our approach reveals many seemingly unrelated estimation schemes as special instances of a general class of variational multiscale estimators, named MIND (MultIscale Nemirovskii--Dantzig). These estimators result from minimizing certain regularization functionals under convex constraints that can be seen as multiple statistical tests for local hypotheses. For computational purposes, we recast MIND in terms of simpler unconstraint optimization problems via Lagrangian penalization as well as Fenchel duality. Performance of several MINDs is demonstrated on numerical examples.

References in corpus (17)

• Simultaneous analysis of Lasso and Dantzig selector
• NESTA: A Fast and Accurate First-order Method for Sparse Recovery
• Adaptive piecewise polynomial estimation via trend filtering
• Can one estimate the conditional distribution of post-model-selection estimators?
• Consistencies and rates of convergence of jump-penalized least squares estimators
• Multiscale local change point detection with applications to value-at-risk
• A data-driven block thresholding approach to wavelet estimation
• Optimal rates for total variation denoising
• Statistical Multiresolution Dantzig Estimation in Imaging: Fundamental Concepts and Algorithmic Framework
• FDR-Control in Multiscale Change-point Segmentation
• Multiscale Change-point Segmentation: Beyond Step Functions
• Extreme Value Analysis of Empirical Frame Coefficients and Implications for Denoising by Soft-Thresholding
• The Essential Histogram
• Empirical Risk Minimization as Parameter Choice Rule for General Linear Regularization Methods
• Variational Multiscale Nonparametric Regression: Smooth Functions
• Calibrating the scan statistic: finite sample performance vs. asymptotics
• Variational Multiscale Nonparametric Regression: Algorithms and Implementation