Confidence sets based on sparse estimators are necessarily large
From MaRDI portal
Abstract: Confidence sets based on sparse estimators are shown to be large compared to more standard confidence sets, demonstrating that sparsity of an estimator comes at a substantial price in terms of the quality of the estimator. The results are set in a general parametric or semiparametric framework.
Recommendations
- Confidence sets in sparse regression
- Estimation and confidence sets for sparse normal mixtures
- Learning Theory
- Sparseness vs estimating conditional probabilities: some asymptotic results
- Fixed-size confidence regions in high-dimensional sparse linear regression models
- A general theory of hypothesis tests and confidence regions for sparse high dimensional models
- Uniformly valid confidence sets based on the Lasso
- Confidence intervals for sparse penalized regression with random designs
- Asymptotic Confidence Regions for High-Dimensional Structured Sparsity
- A new general method for constructing confidence sets in arbitrary dimensions: with applications
Cited in
(19)- On various confidence intervals post-model-selection
- Uniform inference in high-dimensional dynamic panel data models with approximately sparse fixed effects
- On Hodges' superefficiency and merits of oracle property in model selection
- UNIFORM ASYMPTOTICS AND CONFIDENCE REGIONS BASED ON THE ADAPTIVE LASSO WITH PARTIALLY CONSISTENT TUNING
- Uniformly valid confidence intervals post-model-selection
- On the asymptotic variance of the debiased Lasso
- Simultaneous inference for pairwise graphical models with generalized score matching
- Confidence sets in sparse regression
- On the post selection inference constant under restricted isometry properties
- On the length of post-model-selection confidence intervals conditional on polyhedral constraints
- Confidence sets based on penalized maximum likelihood estimators in Gaussian regression
- Inference in Additively Separable Models With a High-Dimensional Set of Conditioning Variables
- Uniformly valid inference based on the Lasso in linear mixed models
- On asymptotically optimal confidence regions and tests for high-dimensional models
- Asymptotically honest confidence regions for high dimensional parameters by the desparsified conservative Lasso
- On the uniform asymptotic validity of subsampling and the bootstrap
- Robust inference on average treatment effects with possibly more covariates than observations
- Model-Assisted Uniformly Honest Inference for Optimal Treatment Regimes in High Dimension
- Confidence intervals that utilize sparsity
This page was built for publication: Confidence sets based on sparse estimators are necessarily large
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q987107)
