ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 1105.1475
129
152

Pivotal estimation via square-root Lasso in nonparametric regression

7 May 2011
A. Belloni
Victor Chernozhukov
Lie Wang
ArXivPDFHTML
Abstract

We propose a self-tuning Lasso\sqrt{\mathrm {Lasso}}Lasso​ method that simultaneously resolves three important practical problems in high-dimensional regression analysis, namely it handles the unknown scale, heteroscedasticity and (drastic) non-Gaussianity of the noise. In addition, our analysis allows for badly behaved designs, for example, perfectly collinear regressors, and generates sharp bounds even in extreme cases, such as the infinite variance case and the noiseless case, in contrast to Lasso. We establish various nonasymptotic bounds for Lasso\sqrt{\mathrm {Lasso}}Lasso​ including prediction norm rate and sparsity. Our analysis is based on new impact factors that are tailored for bounding prediction norm. In order to cover heteroscedastic non-Gaussian noise, we rely on moderate deviation theory for self-normalized sums to achieve Gaussian-like results under weak conditions. Moreover, we derive bounds on the performance of ordinary least square (ols) applied to the model selected by Lasso\sqrt{\mathrm {Lasso}}Lasso​ accounting for possible misspecification of the selected model. Under mild conditions, the rate of convergence of ols post Lasso\sqrt{\mathrm {Lasso}}Lasso​ is as good as Lasso\sqrt{\mathrm {Lasso}}Lasso​'s rate. As an application, we consider the use of Lasso\sqrt{\mathrm {Lasso}}Lasso​ and ols post Lasso\sqrt{\mathrm {Lasso}}Lasso​ as estimators of nuisance parameters in a generic semiparametric problem (nonlinear moment condition or ZZZ-problem), resulting in a construction of n\sqrt{n}n​-consistent and asymptotically normal estimators of the main parameters.

View on arXiv
Comments on this paper