Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386794388> ?p ?o ?g. }
Showing items 1 to 67 of
67
with 100 items per page.
- W4386794388 abstract "We introduce a new debiasing framework for high-dimensional linear regression that bypasses the restrictions on covariate distributions imposed by modern debiasing technology. We study the prevalent setting where the number of features and samples are both large and comparable. In this context, state-of-the-art debiasing technology uses a degrees-of-freedom correction to remove shrinkage bias of regularized estimators and conduct inference. However, this method requires that the observed samples are i.i.d., the covariates follow a mean zero Gaussian distribution, and reliable covariance matrix estimates for observed features are available. This approach struggles when (i) covariates are non-Gaussian with heavy tails or asymmetric distributions, (ii) rows of the design exhibit heterogeneity or dependencies, and (iii) reliable feature covariance estimates are lacking. To address these, we develop a new strategy where the debiasing correction is a rescaled gradient descent step (suitably initialized) with step size determined by the spectrum of the sample covariance matrix. Unlike prior work, we assume that eigenvectors of this matrix are uniform draws from the orthogonal group. We show this assumption remains valid in diverse situations where traditional debiasing fails, including designs with complex row-column dependencies, heavy tails, asymmetric properties, and latent low-rank structures. We establish asymptotic normality of our proposed estimator (centered and scaled) under various convergence notions. Moreover, we develop a consistent estimator for its asymptotic variance. Lastly, we introduce a debiased Principal Component Regression (PCR) technique using our Spectrum-Aware approach. In varied simulations and real data experiments, we observe that our method outperforms degrees-of-freedom debiasing by a margin." @default.
- W4386794388 created "2023-09-16" @default.
- W4386794388 creator A5001559651 @default.
- W4386794388 creator A5067462102 @default.
- W4386794388 date "2023-09-14" @default.
- W4386794388 modified "2023-09-30" @default.
- W4386794388 title "Spectrum-Aware Adjustment: A New Debiasing Framework with Applications to Principal Components Regression" @default.
- W4386794388 doi "https://doi.org/10.48550/arxiv.2309.07810" @default.
- W4386794388 hasPublicationYear "2023" @default.
- W4386794388 type Work @default.
- W4386794388 citedByCount "0" @default.
- W4386794388 crossrefType "posted-content" @default.
- W4386794388 hasAuthorship W4386794388A5001559651 @default.
- W4386794388 hasAuthorship W4386794388A5067462102 @default.
- W4386794388 hasBestOaLocation W43867943881 @default.
- W4386794388 hasConcept C105795698 @default.
- W4386794388 hasConcept C11413529 @default.
- W4386794388 hasConcept C119043178 @default.
- W4386794388 hasConcept C121332964 @default.
- W4386794388 hasConcept C151730666 @default.
- W4386794388 hasConcept C15744967 @default.
- W4386794388 hasConcept C158693339 @default.
- W4386794388 hasConcept C178650346 @default.
- W4386794388 hasConcept C185429906 @default.
- W4386794388 hasConcept C188147891 @default.
- W4386794388 hasConcept C27438332 @default.
- W4386794388 hasConcept C2779343474 @default.
- W4386794388 hasConcept C2779458634 @default.
- W4386794388 hasConcept C28826006 @default.
- W4386794388 hasConcept C33923547 @default.
- W4386794388 hasConcept C41008148 @default.
- W4386794388 hasConcept C62520636 @default.
- W4386794388 hasConcept C86803240 @default.
- W4386794388 hasConceptScore W4386794388C105795698 @default.
- W4386794388 hasConceptScore W4386794388C11413529 @default.
- W4386794388 hasConceptScore W4386794388C119043178 @default.
- W4386794388 hasConceptScore W4386794388C121332964 @default.
- W4386794388 hasConceptScore W4386794388C151730666 @default.
- W4386794388 hasConceptScore W4386794388C15744967 @default.
- W4386794388 hasConceptScore W4386794388C158693339 @default.
- W4386794388 hasConceptScore W4386794388C178650346 @default.
- W4386794388 hasConceptScore W4386794388C185429906 @default.
- W4386794388 hasConceptScore W4386794388C188147891 @default.
- W4386794388 hasConceptScore W4386794388C27438332 @default.
- W4386794388 hasConceptScore W4386794388C2779343474 @default.
- W4386794388 hasConceptScore W4386794388C2779458634 @default.
- W4386794388 hasConceptScore W4386794388C28826006 @default.
- W4386794388 hasConceptScore W4386794388C33923547 @default.
- W4386794388 hasConceptScore W4386794388C41008148 @default.
- W4386794388 hasConceptScore W4386794388C62520636 @default.
- W4386794388 hasConceptScore W4386794388C86803240 @default.
- W4386794388 hasLocation W43867943881 @default.
- W4386794388 hasOpenAccess W4386794388 @default.
- W4386794388 hasPrimaryLocation W43867943881 @default.
- W4386794388 hasRelatedWork W2009565450 @default.
- W4386794388 hasRelatedWork W2022232878 @default.
- W4386794388 hasRelatedWork W2038710811 @default.
- W4386794388 hasRelatedWork W2055894834 @default.
- W4386794388 hasRelatedWork W2460185865 @default.
- W4386794388 hasRelatedWork W3123910301 @default.
- W4386794388 hasRelatedWork W3216626455 @default.
- W4386794388 hasRelatedWork W4246387931 @default.
- W4386794388 hasRelatedWork W4362602876 @default.
- W4386794388 hasRelatedWork W2621420575 @default.
- W4386794388 isParatext "false" @default.
- W4386794388 isRetracted "false" @default.
- W4386794388 workType "article" @default.