Sparse Estimators and the Oracle Property, or the Return of Hodges' Estimator

Abstract

We point out some pitfalls related to the concept of an oracle property as used in Fan and Li (2001, 2002, 2004) which are reminiscent of the well-known pitfalls related to Hodges' estimator. The oracle property is often a consequence of sparsity of an estimator. We show that any estimator satisfying a sparsity property has maximal risk that converges to the supremum of the loss function; in particular, the maximal risk diverges to infinity whenever the loss function is unbounded. For ease of presentation the result is set in the framework of a linear regression model, but generalizes far beyond that setting. In a Monte Carlo study we also assess the extent of the problem in finite samples for the smoothly clipped absolute deviation (SCAD) estimator introduced in Fan and Li (2001). We find that this estimator can perform rather poorly in finite samples and that its worst-case performance relative to maximum likelihood deteriorates with increasing sample size when the estimator is tuned to sparsity.

18 pages, 5 figures

Cited by in corpus (14)

• Consistencies and rates of convergence of jump-penalized least squares estimators
• Evaluation and selection of models for out-of-sample prediction when the sample size is small relative to the complexity of the data-generating process
• Sparsity considerations for dependent observations
• Discussion: "A significance test for the lasso"
• On the Distribution of Penalized Maximum Likelihood Estimators: The LASSO, SCAD, and Thresholding
• The costs and benefits of uniformly valid causal inference with high-dimensional nuisance parameters
• A Generalized Focused Information Criterion for GMM
• On the Distribution of the Adaptive LASSO Estimator
• Confidence intervals for intentionally biased estimators
• Posterior Convergence and Model Estimation in Bayesian Change-point Problems
• Adaptive Ridge Approach to Heteroscedastic Regression
• Understanding the population structure correction regression
• Superconsistency of Tests in High Dimensions
• Minimum Variance Estimation of a Sparse Vector within the Linear Gaussian Model: An RKHS Approach