Aggregate Inverse Mean Estimation for Sufficient Dimension Reduction
- 2 July 2020
- journal article
- research article
- Published by Taylor & Francis in Technometrics
- Vol. 63 (4) , 456-465
Abstract
Many well-known sufficient dimension reduction methods investigate the inverse conditional moments of the predictors given the response. The required linearity condition, the number and arrangement of slices, and the inability to detect symmetric dependence are among several long-standing issues that have negatively impacted on the use of these approaches. Motivated by two recent work dealing with the choice of number of slices, we propose a novel and effective method based on the aggregation of inverse mean estimation (AIME). The new approach can substantially improve the estimation accuracy, break down the symmetry to achieve exhaustive estimation, and is much less sensitive to the violation of the linearity condition. Both simulation studies and a real data application show the efficacy of the newly proposed approach.Keywords
Funding Information
- NSF (CIF-1813330)
This publication has 32 references indexed in Scilit:
- Sufficient dimension reduction through discretization-expectation estimationBiometrika, 2010
- Contour projected dimension reductionThe Annals of Statistics, 2009
- Dimension reduction for nonelliptically distributed predictorsThe Annals of Statistics, 2009
- Sliced Regression for Dimension ReductionJournal of the American Statistical Association, 2008
- A constructive approach to the estimation of dimension reduction directionsThe Annals of Statistics, 2007
- Dimension reduction in regression without matrix inversionBiometrika, 2007
- On Sliced Inverse Regression With High-Dimensional CovariatesJournal of the American Statistical Association, 2006
- Using the Bootstrap to Select One of a New Class of Dimension Reduction MethodsJournal of the American Statistical Association, 2003
- Theory & Methods: Special Invited Paper: Dimension Reduction and Visualization in Discriminant Analysis (with discussion)Australian & New Zealand Journal of Statistics, 2001
- Sliced Inverse Regression for Dimension ReductionJournal of the American Statistical Association, 1991