Estimator Augmentation with Applications in High-Dimensional Group Inference
arXiv:1610.08621 · doi:10.1214/17-EJS1309
Abstract
To make inference about a group of parameters on high-dimensional data, we develop the method of estimator augmentation for the block Lasso, which is defined via the block norm. By augmenting a block Lasso estimator with the subgradient of the block norm evaluated at , we derive a closed-form density for the joint distribution of under a high-dimensional setting. This allows us to draw from an estimated sampling distribution of , or more generally any function of , by Monte Carlo algorithms. We demonstrate the application of estimator augmentation in group inference with the group Lasso and a de-biased group Lasso constructed as a function of . Our numerical results show that importance sampling via estimator augmentation can be orders of magnitude more efficient than parametric bootstrap in estimating tail probabilities for significance tests. This work also brings new insights into the geometry of the sample space and the solution uniqueness of the block Lasso.
arXiv comment35 pages, 4 figures
References in corpus (10)
- On asymptotically optimal confidence regions and tests for high-dimensional models
- Confidence Intervals and Hypothesis Testing for High-Dimensional Regression
- Exact post-selection inference, with application to the lasso
- High-dimensional variable selection
- High-Dimensional Inference: Confidence Intervals, -Values and R-Software hdi
- Rates of convergence of the Adaptive LASSO estimators to the Oracle distribution and higher order refinements by the bootstrap
- Inference in High Dimensions with the Penalized Score Test
- Selective sampling after solving a convex problem
- Rejoinder: "A significance test for the lasso"
- Monte Carlo Simulation for Lasso-Type Problems by Estimator Augmentation