Sharp Threshold Detection Based on Sup-Norm Error Rates in High-Dimensional Models
Bibliographic Data
| ID | 19419355 |
|---|---|
| Authors | Laurent Callot (0000-0002-0756-8120, Department of Econometrics and Operations Research, VU University Amsterdam, CREATES, and the Tinbergen Institute, Amsterdam, The Netherlands()), Mehmet Caner (0000-0002-5566-8476, Department of Economics, Translational Data Analytics, Department of Statistics, 452 Arps Hall, Ohio State University, Columbus, OH 43210 ()), Anders Bredahl Kock (Aarhus University), Anders Kock (Aarhus University), Andrés Riquelme (0000-0003-1220-0926, University of Talca), Juan Andres Riquelme (University of Talca) |
| Year | 2017 |
| Volume | 35 |
| Issue | 2 |
| Pages | 250-264 |
| Publication date | 2017-04-03 |
| Peer Reviewed | Yes |
| Open Access | No |
| Type | ARTICLE |
| Venue | Journal of Business and Economic Statistics (JOURNAL) |
| Journal identifiers | ISSN: 0735-0015 • E-ISSN: 1537-2707 |
| Publisher | Informa UK Limited (PUBLISHER • GB) |
| DOI | 10.1080/07350015.2015.1052461 |
| OpenAlex | W2145437650 |
| Language | EN |
| Citations received | 2 |
| References cited | 24 |
We propose a new estimator, the thresholded scaled Lasso, in high-dimensional threshold regressions. First, we establish an upper bound on the l∞ estimation error of the scaled Lasso estimator of Lee, Seo, and Shin. This is a nontrivial task as the literature on high-dimensional models has focused almost exclusively on l1 and l2 estimation errors. We show that this sup-norm bound can be used to distinguish between zero and nonzero coefficients at a much finer scale than would have been possible using classical oracle inequalities. Thus, our sup-norm bound is tailored to consistent variable selection via thresholding. Our simulations show that thresholding the scaled Lasso yields substantial improvements in terms of variable selection. Finally, we use our estimator to shed further empirical light on the long-running debate on the relationship between the level of debt (public and private) and GDP growth. Supplementary materials for this article are available online
Econometrics · Political science · Statistics · Advanced Statistical Methods and Models · Computer Science · Image and Signal Denoising Methods · Law · Mathematics · Statistical Methods and Inference
Sovereign Debt and the Financial Crisis
Weak Convergence and Empirical Processes
Sample Splitting and Threshold Estimation
Instrumental Variable Estimation of a Threshold Model
Inference When a Nuisance Parameter Is Not Identified Under the Null Hypothesis
Growth in a Time of Debt
Does high public debt consistently stifle economic growth? A critique of Reinhart and Rogoff
Regularization Paths for Generalized Linear Models via Coordinate Descent
Threshold effects in non-dynamic panels
| Unique citing works | 2 |
|---|---|
| Citations per year | 0.5 |
| Citation span | 2022 - 2023 (2) |
| Citation velocity | historical |
| Highly cited | No |
| Citation types | Neutral: 2 |