Adapting to unknown sparsity by controlling the false discovery rate
From MaRDI portal
Publication:2497176
Abstract: We attempt to recover an -dimensional vector observed in white noise, where is large and the vector is known to be sparse, but the degree of sparsity is unknown. We consider three different ways of defining sparsity of a vector: using the fraction of nonzero terms; imposing power-law decay bounds on the ordered entries; and controlling the norm for small. We obtain a procedure which is asymptotically minimax for loss, simultaneously throughout a range of such sparsity classes. The optimal procedure is a data-adaptive thresholding scheme, driven by control of the {it False Discovery Rate} (FDR). FDR control is a relatively recent innovation in simultaneous testing, ensuring that at most a certain fraction of the rejected null hypotheses will correspond to false rejections. In our treatment, the FDR control parameter also plays a determining role in asymptotic minimaxity. If and also we get sharp asymptotic minimaxity, simultaneously, over a wide range of sparse parameter spaces and loss functions. On the other hand, , forces the risk to exceed the minimax risk by a factor growing with . To our knowledge, this relation between ideas in simultaneous inference and asymptotic decision theory is new. Our work provides a new perspective on a class of model selection rules which has been introduced recently by several authors. These new rules impose complexity penalization of the form . We exhibit a close connection with FDR-controlling procedures under stringent control of the false discovery rate.
Recommendations
- Asymptotic minimaxity of false discovery rate thresholding for sparse exponential data
- On false discovery rate thresholding for classification under sparsity
- Some optimality properties of FDR controlling rules under sparsity
- Asymptotic Bayes-optimality under sparsity of some multiple testing procedures
- A simple forward selection procedure based on false discovery rate control
Cites work
- A Direct Approach to False Discovery Rates
- A stagewise rejective multiple test procedure based on a modified Bonferroni test
- A stochastic process approach to false discovery control.
- A wavelet auditory model and data compression
- Adapting to Unknown Smoothness via Wavelet Shrinkage
- Adapting to unknown sparsity by controlling the false discovery rate
- Adaptive linear step-up procedures that control the false discovery rate
- An improved Bonferroni procedure for multiple tests of significance
- Calibration and empirical Bayes variable selection
- Convergence of stochastic processes
- De-noising by soft-thresholding
- Empirical Bayes Analysis of a Microarray Experiment
- Estimating a regression function
- Estimating the dimension of a model
- Gaussian model selection
- How Biased is the Apparent Error Rate of a Prediction Rule?
- How Many Variables Should be Entered in a Regression Equation?
- scientific article; zbMATH DE number 50804 (Why is no real title available?)
- scientific article; zbMATH DE number 409717 (Why is no real title available?)
- scientific article; zbMATH DE number 524360 (Why is no real title available?)
- scientific article; zbMATH DE number 720689 (Why is no real title available?)
- scientific article; zbMATH DE number 724718 (Why is no real title available?)
- scientific article; zbMATH DE number 739533 (Why is no real title available?)
- scientific article; zbMATH DE number 3444596 (Why is no real title available?)
- scientific article; zbMATH DE number 802850 (Why is no real title available?)
- scientific article; zbMATH DE number 1405825 (Why is no real title available?)
- Image compression through wavelet transform coding
- Local asymptotic coding and the minimum description length
- Minimax risk over \(l_ p\)-balls for \(l_ q\)-error
- Needles and straw in haystacks: Empirical Bayes estimates of possibly sparse sequences
- Some Comments on C P
- Some results on false discovery rate in stepwise multiple testing procedures.
- Strong Control, Conservative Point Estimation and Simultaneous Conservative Consistency of False Discovery Rates: A Unified Approach
- The control of the false discovery rate in multiple testing under dependency.
- The Covariance Inflation Criterion for Adaptive Model Selection
- The risk inflation criterion for multiple regression
- The statistics of natural images
Cited in
(only showing first 100 items - show all)- Covariance regularization by thresholding
- A universal procedure for aggregating estimators
- A simple forward selection procedure based on false discovery rate control
- Control of the false discovery rate under dependence using the bootstrap and subsampling
- On the false discovery rate and an asymptotically optimal rejection curve
- An adaptive step-down procedure with proven FDR control under independence
- Gaussian model selection with an unknown variance
- False Discovery Rate Control Under General Dependence By Symmetrized Data Aggregation
- Thresholding rules for recovering a sparse signal from microarray experiments
- Consistent change-point detection with kernels
- A multiple testing approach to the regularisation of large sample correlation matrices
- An adaptation theory for nonparametric confidence intervals
- General empirical Bayes wavelet methods and exactly adaptive minimax estimation
- Variable selection with Hamming loss
- Some results on false discovery rate in stepwise multiple testing procedures.
- Higher criticism for detecting sparse heterogeneous mixtures.
- A stochastic process approach to false discovery control.
- Least angle regression. (With discussion)
- Minimax estimation of linear functionals over nonconvex parameter spaces.
- Needles and straw in haystacks: Empirical Bayes estimates of possibly sparse sequences
- FDR-controlling stepwise procedures and their false negatives rates
- Needles and straw in a haystack: posterior concentration for possibly sparse sequences
- Model selection in regression under structural constraints
- Some optimality properties of FDR controlling rules under sparsity
- Data-driven wavelet-Fisz methodology for nonparametric function estimation
- MAP model selection in Gaussian regression
- Ideal denoising within a family of tree-structured wavelet estimators
- Model selection and sharp asymptotic minimaxity
- Comparison of Bayesian and frequentist multiplicity correction for testing mutually exclusive hypotheses under data dependence
- Rate of convergence of a risk estimator to the normal law in a multiple hypothesis testing problem using the FDR threshold
- Reproducible feature selection in high-dimensional accelerated failure time models
- Reproducible learning in large-scale graphical models
- Uncertainty quantification for robust variable selection and multiple testing
- Rate of convergence of the risk estimate distribution to the normal law using FDR multiple hypothesis testing with inverting linear homogeneous operators
- Distributed nonparametric function estimation: optimal rate of convergence and cost of adaptation
- GSDAR: a fast Newton algorithm for \(\ell_0\) regularized generalized linear models with statistical guarantee
- TFisher: a powerful truncation and weighting procedure for combining \(p\)-values
- Asymptotically optimal sequential FDR and pFDR control with (or without) prior information on the number of signals
- On the asymptotic properties of SLOPE
- On spike and slab empirical Bayes multiple testing
- Sparse semiparametric discriminant analysis
- On signal reconstruction in white noise using dictionaries
- Unsupervised empirical Bayesian multiple testing with external covariates
- Modified versions of the Bayesian information criterion for sparse generalized linear models
- Spike and slab empirical Bayes sparse credible sets
- Minimal penalties for Gaussian model selection
- Asymptotic minimaxity of false discovery rate thresholding for sparse exponential data
- A two-sample test for high-dimensional data with applications to gene-set testing
- General maximum likelihood empirical Bayes estimation of normal means
- Good, great, or lucky? Screening for firms with sustained superior performance using heavy-tailed priors
- A guided random walk through some high dimensional problems
- Sparse PCA: optimal rates and adaptive estimation
- Sparse model selection under heterogeneous noise: exact penalisation and data-driven thresholding
- On the performance of FDR control: constraints and a partial solution
- On optimality of Bayesian testimation in the normal means problem
- The false discovery rate: a variable selection perspective
- Adapting to unknown sparsity by controlling the false discovery rate
- Consistent variable selection in high dimensional regression via multiple testing
- Empirical Bayes selection of wavelet thresholds
- Minimax and adaptive inference in nonparametric function estimation
- Simultaneous testing of multiple hypotheses using generalized \(p\)-values
- Strong consistency of the risk estimator in multiple hypothesis testing with the FDR threshold
- On the optimality of sliced inverse regression in high dimensions
- Consistent variable selection for functional regression models
- Sharp variable selection of a sparse submatrix in a high-dimensional noisy matrix
- SLOPE is adaptive to unknown sparsity and asymptotically minimax
- Higher criticism thresholding: Optimal feature selection when useful features are rare and weak
- Fast rate of convergence in high-dimensional linear discriminant analysis
- Combining thresholding rules: A new way to improve the performance of wavelet estimators
- Functional mixed effects wavelet estimation for spectra of replicated time series
- Estimation of matrices with row sparsity
- RANK: Large-Scale Inference With Graphical Nonlinear Knockoffs
- Generalized Augmentation to Control the False Discovery Exceedance in Multiple Testing
- Feature selection by higher criticism thresholding achieves the optimal phase diagram
- Adjusting Stepwisep-Values in Generalized Linear Models
- A review of modern multiple hypothesis testing, with particular attention to the false discovery proportion
- Proportion of Non-Zero Normal Means: Universal Oracle Equivalences and Uniformly Consistent Estimators
- A nonparametric empirical Bayes approach to adaptive minimax estimation
- Feature selection when there are many influential features
- On the Benjamini-Hochberg method
- UPS delivers optimal phase diagram in high-dimensional variable selection
- Controlling the false discoveries in Lasso
- Controlling the familywise error rate with plug-in estimator for the proportion of true null hypotheses
- Sparse partial least squares regression for simultaneous dimension reduction and variable selection
- Discovering the false discovery rate. With comments and author's reply
- Regularization and the small-ball method. II: Complexity dependent error rates
- Operating Characteristics and Extensions of the False Discovery Rate Procedure
- Asymptotically minimax empirical Bayes estimation of a sparse normal mean vector
- Nonparametric significance testing and group variable selection
- Binary classification with pFDR-pFNR losses
- Minimax optimal estimation of high-dimensional sparse covariance matrices with missing data
- Covariance-engaged classification of sets via linear programming
- Adaptive sparse estimation with side information
- Local comparison of empirical distributions via nonparametric regression
- Group SLOPE – Adaptive Selection of Groups of Predictors
- Hierarchical recognition of sparse patterns in large-scale simultaneous inference
- Local posterior concentration rate for multilevel sparse sequences
- Block-threshold-adapted estimators via a maxiset approach
- Model selection for estimating the non zero components of a Gaussian vector
- Exponential screening and optimal rates of sparse estimation
This page was built for publication: Adapting to unknown sparsity by controlling the false discovery rate
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2497176)
