Please use this identifier to cite or link to this item:
Title: High-dimensional influence measure
Authors: Zhao, J.
Leng, C. 
Li, L.
Wang, H.
Keywords: Cook's distance
High-dimensional diagnosis
Influential observation
Marginal correlations
Variable screening
Issue Date: Oct-2013
Citation: Zhao, J., Leng, C., Li, L., Wang, H. (2013-10). High-dimensional influence measure. Annals of Statistics 41 (5) : 2639-2667. ScholarBank@NUS Repository.
Abstract: Influence diagnosis is important since presence of influential observations could lead to distorted analysis and misleading interpretations. For highdimensional data, it is particularly so, as the increased dimensionality and complexity may amplify both the chance of an observation being influential, and its potential impact on the analysis. In this article, we propose a novel high-dimensional influence measure for regressions with the number of predictors far exceeding the sample size. Our proposal can be viewed as a highdimensional counterpart to the classical Cook's distance. However, whereas the Cook's distance quantifies the individual observation's influence on the least squares regression coefficient estimate, our new diagnosis measure captures the influence on the marginal correlations, which in turn exerts serious influence on downstream analysis including coefficient estimation, variable selection and screening. Moreover, we establish the asymptotic distribution of the proposed influence measure by letting the predictor dimension go to infinity. Availability of this asymptotic distribution leads to a principled rule to determine the critical value for influential observation detection. Both simulations and real data analysis demonstrate usefulness of the new influence diagnosis measure. © Institute of Mathematical Statistics, 2013.
Source Title: Annals of Statistics
ISSN: 00905364
DOI: 10.1214/13-AOS1165
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

Google ScholarTM



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.