Minimum Average Deviance Estimation for Sufficient Dimension Reduction
Authors:
Preprint
Abstract
Sufficient dimension reduction reduces the dimensionality of data while preserving relevant regression information. In this article, we develop Minimum Average Deviance Estimation (MADE) methodology for sufficient dimension reduction. It extends the Minimum Average Variance Estimation (MAVE) approach of Xia et al. (2002) from continuous responses to exponential family distributions to include Binomial and Poisson responses. Local likelihood regression is used to learn the form of the regression function from the data. The main parameter of interest is a dimension reduction subspace which projects the covariates to a lower dimension while preserving their relationship with the outcome. To estimate this parameter within its natural space, we consider an iterative algorithm where one step utilizes a Stiefel manifold optimizer. We empirically evaluate the performance of three prediction methods, two that are intrinsic to local likelihood estimation and one that is based on the Nadaraya-Watson estimator. Initial results show that, as expected, MADE can outperform MAVE when there is a departure from the assumption of additive errors.Keywords
All Related Versions
This publication has 6 references indexed in Scilit:
- Penalized minimum average variance estimationStatistica Sinica, 2013
- Optimization Algorithms on Matrix Manifolds2008
- On Directional Regression for Dimension ReductionJournal of the American Statistical Association, 2007
- Nonlinear regression modeling using regularized local likelihood methodAnnals of the Institute of Statistical Mathematics, 2005
- Contour regression: A general approach to dimension reductionThe Annals of Statistics, 2005
- Local Polynomial Modeling and Its ApplicationsJournal of the American Statistical Association, 1998