arXiv 20 Mar 2019 · Mathematics — Statistics Theory · publishedThe Review of Economics and Statistics (2021) · 5 citations (OpenAlex)
arXiv:1903.08704 · PDF · DOI · OpenAlex · Extracted main text
We study the finite sample behavior of Lasso-based inference methods such as post double Lasso and debiased Lasso. We show that these methods can exhibit substantial omitted variable biases (OVBs) due to Lasso not selecting relevant controls. This phenomenon can occur even when the coefficients are sparse and the sample size is large and larger than the number of controls. Therefore, relying on the existing asymptotic inference theory can be problematic in empirical applications. We compare the Lasso-based inference methods to modern high-dimensional OLS-based methods and provide practical guidance.
appendix boundary found by appendix_command · 43% of the source is main text. Read the extracted text to check this.
The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.
| Reference | Intensity | Mentions | Sections | Main text | |
|---|---|---|---|---|---|
| 1 | Alexandre Belloni, Daniel Chen, Victor Chernozhukov, and Christian H… (2012) Sparse models and methods for optimal instruments with an application to eminent domain | 1.000 | 7 | 4 | 100% |
| 2 | Peter J. Bickel, Ya'acov Ritov, and Alexandre B. Tsybakov Simultaneous analysis of lasso and dantzig selector | 1.000 | 6 | 4 | 100% |
| 3 | Alexandre Belloni, Victor Chernozhukov, Iván Fernández-Val, and Chri… (2017) Program evaluation and causal inference with high-dimensional data | 1.000 | 6 | 3 | 100% |
| 4 | Alexandre Belloni, Victor Chernozhukov, and Christian Hansen (2014) Inference on treatment effects after selection among high-dimensional controls | 0.956 | 24 | 9 | 88% |
| 5 | Victor Chernozhukov, Denis Chetverikov, Mert Demirer, Esther Duflo,… (2018) Double/debiased machine learning for treatment and structural parameters | 0.928 | 4 | 3 | 100% |
| 6 | Martin J. Wainwright (2009) Sharp thresholds for high-dimensional and noisy sparsity recovery using $ _1$-constrained quadratic programming (lasso) | 0.920 | 9 | 4 | 78% |
| 7 | Matias D. Cattaneo, Michael Jansson, and Whitney K. Newey (2018) Inference in linear regression models with many covariates and heteroscedasticity | 0.874 | 5 | 2 | 100% |
| 8 | Joshua D. Angrist and Brigham Frandsen (2019) Machine labor | 0.843 | 4 | 4 | 75% |
| 9 | Soumendra N. Lahiri (2021) Necessary and sufficient conditions for variable selection consistency of the lasso in high dimensions | 0.811 | 4 | 2 | 100% |
| 10 | Roland G. Fryer and Steven D. Levitt (2013) Testing for racial differences in the mental ability of young children | 0.737 | 3 | 2 | 100% |
Showing the top 10 of 65 scored citations.
arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.