Asymptotics for sliced average variance estimation

DOI10.1214/009053606000001091MaRDI QIDQ997370zbMATH OpenOpenAlexFDO

Authors Xing-Xing Li, Li-Xing Zhu

Publication date 23 July 2007

Published in The Annals of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/0708.0462

zbMATH Keywords

convergence ratedimension reductionsliced average variance estimation

Mathematics Subject Classification ID

Asymptotic distribution theory in statistics (62E20) Nonparametric regression and quantile regression (62G08) Asymptotic properties of nonparametric inference (62G20) Estimation in multivariate analysis (62H12) Multivariate analysis (62H99)

Abstract: In this paper, we systematically study the consistency of sliced average variance estimation (SAVE). The findings reveal that when the response is continuous, the asymptotic behavior of SAVE is rather different from that of sliced inverse regression (SIR). SIR can achieve $sqrtn$ consistency even when each slice contains only two data points. However, SAVE cannot be $sqrtn$ consistent and it even turns out to be not consistent when each slice contains a fixed number of data points that do not depend on n, where n is the sample size. These results theoretically confirm the notion that SAVE is more sensitive to the number of slices than SIR. Taking this into account, a bias correction is recommended in order to allow SAVE to be $sqrtn$ consistent. In contrast, when the response is discrete and takes finite values, $sqrtn$ consistency can be achieved. Therefore, an approximation through discretization, which is commonly used in practice, is studied. A simulation study is carried out for the purposes of illustration.

Recommendations

• On kernel method for sliced average variance estimation
• On splines approximation for sliced average variance estimation
• Save: a method for dimension reduction and graphics in regression
• Dimension reduction in regressions through weighted variance estimation
• Asymptotics for kernel estimation of slicing average third-moment estimation

Cites work

• An Adaptive Estimation of Dimension Reduction Space
• An asymptotic theory for sliced inverse regression
• Approximation Theorems of Mathematical Statistics
• Asymptotics for kernel estimate of sliced inverse regression
• Convergence of stochastic processes
• Dimension reduction for conditional mean in regression
• Estimating Optimal Transformations for Multiple Regression and Correlation
• scientific article; zbMATH DE number 1713116 (Why is no real title available?)
• scientific article; zbMATH DE number 1220060 (Why is no real title available?)
• scientific article; zbMATH DE number 1932857 (Why is no real title available?)
• scientific article; zbMATH DE number 788275 (Why is no real title available?)
• Identifying Regression Outliers and Mixtures Graphically
• Model checks for regression: an innovation process approach
• Nonlinear time series. Nonparametric and parametric methods
• Nonparametric checks for single-index models
• On hybrid methods of inverse regression-based algorithms
• On Principal Hessian Directions for Data Visualization and Dimension Reduction: Another Application of Stein's Lemma
• On Sliced Inverse Regression With High-Dimensional Covariates
• On the Interpretation of Regression Plots
• Save: a method for dimension reduction and graphics in regression
• Simultaneous Equations and Canonical Correlation Theory
• Sliced Inverse Regression for Dimension Reduction
• Sufficient Dimension Reduction via Inverse Regression
• Using the Bootstrap to Select One of a New Class of Dimension Reduction Methods

Cited in

(53)

• Partial central subspace and sliced average variance estimation
• Dimension reduction based on weighted variance estimate
• Contour projected dimension reduction
• Nonlinear surface regression with dimension reduction method
• The hybrid method of FSIR and FSAVE for functional effective dimension reduction
• Fused sliced average variance estimation
• Sufficient dimension reduction in regressions through cumulative Hessian directions
• On determining the structural dimension via directional regression
• Advanced topics in sliced inverse regression
• Distributed estimation in heterogeneous reduced rank regression: with application to order determination in sufficient dimension reduction
• Generalized kernel-based inverse regression methods for sufficient dimension reduction
• An ensemble of inverse moment estimators for sufficient dimension reduction
• On kernel method for sliced average variance estimation
• A graphical tool for selecting the number of slices and the dimension of the model in SIR and SAVE approaches
• On the extension of sliced average variance estimation to multivariate regression
• An empirical process view of inverse regression
• Penalized Weighted Variance Estimate for Dimension Reduction
• Partially linear estimation using sufficient dimension reduction
• Dimension reduction in regressions through weighted variance estimation
• A data-adaptive hybrid method for dimension reduction
• Bagging Versions of Sliced Inverse Regression
• Multi-index regression models with missing covariates at random
• General directional regression
• Series expansion for functional sufficient dimension reduction
• An alternating determination-optimization approach for an additive multi-index model
• Principal support vector machines for linear and nonlinear sufficient dimension reduction
• Dimension reduction for the conditional kth moment via central solution space
• On Partial Sufficient Dimension Reduction With Applications to Partially Linear Multi-Index Models
• Dimension reduction via adaptive slicing
• Distributed Sufficient Dimension Reduction for Heterogeneous Massive Data
• Sliced Independence Test
• Variable importance assessment in sliced inverse regression for variable selection
• SAVE: Robust or not?
• Sliced average variance estimation for multivariate time series
• Computational Outlier Detection Methods in Sliced Inverse Regression
• Recursive kernel estimator in a semiparametric regression model
• Variable-dependent partial dimension reduction
• Strong consistency of kernel method for sliced average variance estimation
• Missing data analysis with sufficient dimension reduction
• Inference for the Dimension of a Regression Relationship Using Pseudo-Covariates
• Projection divergence in the reproducing kernel Hilbert space: asymptotic normality, block-wise and slicing estimation, and computational efficiency
• Coordinate-independent sparse sufficient dimension reduction and variable selection
• Data-driven slicing for dimension reduction in regressions: A likelihood-ratio approach
• Sliced average variance estimation for tensor data
• On the structural dimension of sliced inverse regression
• Smote enhanced cumulative moment estimation for sufficient dimension reduction
• Detect complete dependence via trace correlation in the presence of matrix-valued random objects
• Slice weighted average regression
• On expectile-assisted inverse regression estimation for sufficient dimension reduction
• Estimation for a partial-linear single-index model
• Functional sufficient dimension reduction: convergence rates and multiple functional case
• Dimension reduction using the generalized gradient direction
• On splines approximation for sliced average variance estimation

This page was built for publication: Asymptotics for sliced average variance estimation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q997370)