Abstract
We provide comments on the article “High-dimensional simultaneous inference with the bootstrap” by Ruben Dezeure, Peter Buhlmann and Cun-Hui Zhang.
Similar content being viewed by others
Notes
In 1950s, many statisticians including Jack Kiefer, Herbert Robbins, Jack Wolfowitz, R. C. Bose, Abraham Wald and Milton Sobel thought that the area of ranking and selection is a “revolution”.
In paper (2010), Chatterjee and Lahiri showed that bootstrapping the Lasso is inconsistent whenever one or more regression coefficients are zero. They proposed bootstrapping the thresholded Lasso (Chatterjee and Lahiri 2011) and bootstrapping the adaptive Lasso (Chatterjee and Lahiri 2013) and showed their validation under the beta-min condition and other appropriate conditions. However, bootstrapping the Lasso performs comparably (but is simpler) to the other two methods in simulations where the signal-to-noise ratio is not high.
References
Bahadur RR (1950) On the problem in the theory of \(k\) populations. Ann Math Stat 21:362–365
Bahadur RR, Robbins H (1950) The problem of the greater mean. Ann Math Stat 21:469–487
Bechhofer RE (1954) A single-sample multiple decision procedure for ranking means of normal populations with known variances. Ann Math Stat 25:16–39
Bechhofer RE, Kiefer J, Sobel M (1968) Sequential identification and ranking procedures. University of Chicago Press, Chicago
Chatterjee A, Lahiri SN (2010) Asymptotic properties of the residual bootstrap for Lasso estimators. Proc Am Math Soc 138:4497–4509
Chatterjee A, Lahiri SN (2011) Bootstrapping Lasso estimators. J Am Stat Assoc 106:608–625
Chatterjee A, Lahiri SN (2013) Rates of convergence of the adaptive LASSO estimators to the oracle distribution and higher order refinements by the bootstrap. Ann Stat 41:1232–1259
Dezeure R, Bühlmann P, Meier L, Meinshausen N (2014) High-dimensional inference: confidence intervals, \(p\) values and R-software HDI. Statist Sci 30:533–558
Dezeure R, Bühlmann P, Zhang C-H (2017) High-dimensional simultaneous inference with the bootstrap. Test. doi:10.1007/s11749-017-0554-2
Kay KN, Naselaris T, Prenger RJ, Gallant JL (2008) Identifying natural images from human brain activity. Nature 452:352–355
Liu H, Yu B (2013) Asymptotic properties of Lasso \(+\) mLS and Lasso \(+\) ridge in sparse high-dimensional linear regression. Electron J Stat 7:3124–3169
Lu X, Gamst A, Xu R (2009) Rdcurve: a non-parametric method to evaluate the stability of ranking procedures. IEEE/ACM Trans Comput Biol Bioinform. doi:10.1109/TCBB.2008.138
Mukhopadhyay N (2000) A conversation with Milton Sobel. Stat Sci 15(2):168–190
Acknowledgements
We would like to thank the Editor for the invitation to discuss and thank the Gallant Lab at UC Berkeley for providing the fMRI data. We also thank Jasjeet Sekhon for helpful discussions and for comments that clarify the text and thank Rebecca Barter for extremely helpful comments that led to much improvement of the paper.
Author information
Authors and Affiliations
Corresponding author
Additional information
This work is supported in part by NSF Grant DMS-1228246, ONR Grant N00014-16-1-2664, AFOSR Grant FA9550-14-1-0016, and by the Center for Science of Information (CSoI), an NSF Science and Technology Center, under Grant Agreement CCF-0939370 (to Yu).
This comment refers to the invited paper available at: doi:10.1007/s11749-017-0554-2.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Liu, H., Yu, B. Comments on: High-dimensional simultaneous inference with the bootstrap. TEST 26, 740–750 (2017). https://doi.org/10.1007/s11749-017-0559-x
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11749-017-0559-x
Keywords
- Ranking and selection
- Mean squared error
- Coverage
- High-dimensional statistical inference
- Bootstrapping LassoOLS