Sliced Inverse Regression for Dimension Reduction
From MaRDI portal
\(y\) est la variable scalaire à expliquer, \(x\) est le vecteur de dimension \(p\) des variables explicatives. Si le modèle suivant: \[ y=f(\beta_ 1 x,\dots,\beta_ K x,u)\tag{1} \] est valide (où les \(\beta_ k\) sont des \(p\)-vecteurs de paramètres, \(u\) une perturbation aléatoire, \(f\) une fonction arbitraire inconnue: \(\mathbb{R}^{K+1}\to \mathbb{R}\)) celà signifie qu'un espace de dimension \(K<p\) ``capte toute l'information utile de \(x\) vis-à-vis de \(y\). La régression inverse consiste à étudier l'espérance conditionnelle \(E(x\mid y)\). Le résultat central s'appuye sur la condition suivante (C): Pour tout \(b\in \mathbb{R}^ p\), \(E(bx\mid\beta_ 1 x,\dots,\beta_ K x)\) est linéaire en les \(\beta_ k x\). (1) et (C) impliquent que \(E(x\mid y)\) sera contenu dans un espace de dimension \(K\). Pratiquement, cette régression inverse peut s'opérer en découpant ``en tranches l'étendue des valeurs prises par \(y\), puis en déterminant les points moyens des sous-échantillons ainsi obtenus et en effectuant une analyse en composantes principales de ces points moyens (en tenant compte des effectifs qu'ils représentent), les \(K\) premiers facteurs étant seuls retenus. Des simulations numériques sont présentées ainsi que, à la suite de l'article, des commentaires de R. D. Cook, S. Weisberg, D. R. Brillinger, W. Haerdle et A. B. Tsybakov, J. T. Kent.
Recommendations
Cited in
(only showing first 100 items - show all)- Nonconcave penalized inverse regression in single-index models with high dimensional predic\-tors
- New approaches to model-free dimension reduction for bivariate regression
- Partial central subspace and sliced average variance estimation
- Robust estimation of dimension reduction space
- Efficient estimation in sufficient dimension reduction
- Dimension reduction for nonelliptically distributed predictors
- Consistent estimation of the dimensionality in sliced inverse regression
- On hybrid methods of inverse regression-based algorithms
- Class prediction and gene selection for DNA microarrays using regularized sliced inverse regression
- An RKHS formulation of the inverse regression dimension-reduction problem
- Sliced inverse regression for multivariate response regression
- Sliced mean variance-covariance inverse regression
- A novel moment-based sufficient dimension reduction approach in multivariate regression
- A nonlinear multi-dimensional variable selection method for high dimensional data: sparse MAVE
- Trimming influential observations for improved single-index model estimated sufficient summary plots
- An integral transform method for estimating the central mean and central subspaces
- Dimension reduction based on weighted variance estimate
- Contour projected dimension reduction
- Statistical inference on parametric part for partially linear single-index model
- Leveraging mixed and incomplete outcomes via reduced-rank modeling
- Response dimension reduction for the conditional mean in multivariate regression
- Estimating sufficient reductions of the predictors in abundant high-dimensional regressions
- A simple measure of conditional dependence
- A constructive approach to the estimation of dimension reduction directions
- Dimension-reduction type test for linearity of a stochastic regression model
- Exploring regression structure with graphics
- Nonlinear confounding in high-dimensional regression
- On average derivative quantile regression
- Choice of regressors in nonparametric estimation
- A chi-square test for dimensionality with non-Gaussian data
- A quasi-residuals method in sliced inverse regression.
- Bayesian backfitting. (With comments and a rejoinder).
- Dimension reduction for the conditional mean in regressions with categorical predictors
- Sparse sliced inverse regression via Lasso
- Statistical estimation in varying coefficient models
- Semiparametric mixtures of regressions with single-index for model based clustering
- Dimension reduction strategies for analyzing global gene expression data with a response
- Entropy-based sliced inverse regression
- Estimation and empirical likelihood for single-index models with missing data in the covariates
- Bayesian analysis of generalized partially linear single-index models
- Dimension reduction with missing response at random
- Functional \(k\)-means inverse regression
- Dimension reduction based on conditional multiple index density function
- Estimation in linear regression models with measurement errors subject to single-indexed distortion
- A new sliced inverse regression method for multivariate response
- Transformation-based estimation
- A sequential test for variable selection in high dimensional complex data
- Dimension reduction and estimation in the secondary analysis of case-control studies
- On a new class of sufficient dimension reduction estimators
- Nonparametric independence screening via favored smoothing bandwidth
- A unified approach to sufficient dimension reduction
- Quantile-slicing estimation for dimension reduction in regression
- Penalized principal logistic regression for sparse sufficient dimension reduction
- Sufficient dimension reduction constrained through sub-populations
- Principal quantile regression for sufficient dimension reduction with heteroscedasticity
- Student sliced inverse regression
- Sufficient dimension reduction using Hilbert-Schmidt independence criterion
- Partial projective resampling method for dimension reduction: with applications to partially linear models
- Inference for biased transformation models
- Robust estimation and variable selection in sufficient dimension reduction
- Canonical kernel dimension reduction
- Data-driven algorithms for dimension reduction in causal inference
- Ensemble sufficient dimension folding methods for analyzing matrix-valued data
- Heteroscedasticity testing for regression models: a dimension reduction-based model adaptive approach
- Clustering, classification, discriminant analysis, and dimension reduction via generalized hyperbolic mixtures
- Principal minimax support vector machine for sufficient dimension reduction with contaminated data
- On dual model-free variable selection with two groups of variables
- Fused mean-variance filter for feature screening
- Supervised dimension reduction for ordinal predictors
- On sufficient dimension reduction with missing responses through estimating equations
- Estimation and inference on central mean subspace for multivariate response data
- The hybrid method of FSIR and FSAVE for functional effective dimension reduction
- Fused sliced average variance estimation
- Independent component analysis for tensor-valued data
- B spline variable selection for the single index models
- Inverse regression approach to robust nonlinear high-to-low dimensional mapping
- Functional envelope for model-free sufficient dimension reduction
- A relative error estimation approach for multiplicative single index model
- Estimating a sparse reduction for general regression in high dimensions
- Latent single-index models for ordinal data
- Using DAGs to identify the sufficient dimension reduction in the principal fitted components model
- A novel partial-linear single-index model for time series data
- Stability approach to selecting the number of principal components
- Principal weighted logistic regression for sufficient dimension reduction in binary classification
- Tail dimension reduction for extreme quantile estimation
- The effect of data contamination in sliced inverse regression and finite sample breakdown point
- Supervised dimensionality reduction via distance correlation maximization
- On consistency and sparsity for sliced inverse regression in high dimensions
- Dimension reduction-based significance testing in nonparametric regression
- Slice inverse regression with score functions
- Functional multi-layer perceptron: A nonlinear tool for functional data analysis
- A robust adaptive-to-model enhancement test for parametric single-index models
- Linearity identification for general partial linear single-index models
- Joint sufficient dimension reduction for estimating continuous treatment effect functions
- Generalized principal Hessian directions for mixture multivariate skew elliptical distributions
- Quasi-likelihood estimation of the single index conditional variance model
- Semiparametric double robust and efficient estimation for mean functionals with response missing at random
- Robust variable selection through MAVE
- Sufficient dimension reduction in multivariate regressions with categorical predictors
- Two step composite quantile regression for single-index models
This page was built for publication: Sliced Inverse Regression for Dimension Reduction
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3981136)
