Abstract
This work is concerned with testing the marginal linear effects of high-dimensional predictors in quantile regression. We introduce a novel test that is constructed using maxima of pairwise quantile correlations, which permit consistent assessment of the marginal linear effects. The proposed testing procedure is computationally efficient with the aid of a simple multiplier bootstrap method and does not involve any need to select tuning parameters, apart from the number of bootstrap replications. Other distinguishing features of the new procedure are that it imposes no structural assumptions on the unknown dependence structures of the predictor vector and allows the dimension of the predictor vector to be exponentially larger than sample size. To broaden the applicability, we further extend the preceding analysis to the censored response case. The effectiveness of our proposed approach in the finite samples is illustrated through simulation studies.
Similar content being viewed by others
References
Arias-Castro, E., Candès, E. J., Plan, Y. (2011). Global testing under sparse alternatives: ANOVA, multiple comparisons and the higher criticism. The Annals of Statistics, 39, 2533–2556.
Belloni, A., Chernozhukov, V., Kato, K. (2015). Uniform post-selection inference for least absolute deviation regression and other Z-estimation problems. Biometrika, 102, 77–94.
Cai, T. T., Liu, W., Xia, Y. (2014). Two-sample test of high dimensional means under dependence. Journal of the Royal Statistical Society, Series B, 76, 349–372.
Chernozhukov, V., Chetverikov, D., Kato, K. (2013). Gaussian approximations and multiplier bootstrap for maxima of sums of high-dimensional random vectors. The Annals of Statistics, 41, 2786–2819.
Chernozhukov, V., Chetverikov, D., Kato, K. (2015). Comparison and anti-concentration bounds for maxima of Gaussian random vectors. Probability Theory and Related Fields, 162, 47–70.
Chernozhukov, V., Chetverikov, D., Kato, K. (2017). Central limit theorems and bootstrap in high dimensions. The Annals of Probability, 45, 2309–2352.
Conde-Amboage, M., Sánchez-Sellero, C., González-Manteiga, W. (2015). A lack-of-fit test for quantile regression models with high-dimensional covariates. Computational Statistics and Data Analysis, 88, 128–138.
Einmahl, U., Li, D. (2008). Characterization of lil behavior in Banach space. Transactions of the American Mathematical Society 360, 6677–6693.
Escanciano, J. C. (2006). A consistent diagnostic test for regression models using projections. Econometric Theory, 22, 1030–1051.
Hall, P., Jin, J. (2010). Innovated higher criticism for detecting sparse signals in correlated noise. The Annals of Statistics, 38, 1686–1732.
Hall, P., Sheather, S. J. (1988). On the distribution of a studentized quantile. Journal of the Royal Statistical Society, Series B, 50, 381–391.
He, X., Zhu, L. (2003). A lack-of-fit test for quantile regression. Journal of the American Statistical Association, 98, 1013–1022.
He, X., Wang, L., Hong, H. G. (2013). Quantile-adaptive model-free variable screening for high-dimensional heterogeneous data. The Annals of Statistics, 41, 342–369.
Ingster, Y. I., Tsybakov, A. B., Verzelen, N. (2010). Detection boundary in sparse regression. Electronic Journal of Statistics, 4, 1476–1526.
Javanmard, A., Montanari, A. (2014). Confidence intervals and hypothesis testing for high-dimensional regression. Journal of Machine Learning Research, 15, 2869–2909.
Koenker, R. (2005).Quantile regression, Cambridge University Press, New York).
Koenker, R., Bassett, G. (1978). Regression quantiles. Econometrica, 46, 33–50.
Li, G., Li, Y., Tsai, C. L. (2015). Quantile correlations and quantile autoregressive modeling. Journal of the American Statistical Association, 110, 246–261.
Liu, W., Shao, Q. (2013). A Cramér moderate deviation theorem for Hotelling’s \(T^2\)-statistic with applications to global tests. The Annals of Statistics, 41, 296–322.
Liu, W., Lin, Z., Shao, Q. (2008). The asymptotic distribution and Berry-Esseen bound of a new test for independence in high dimension with an application to stochastic optimization. The Annals of Applied Probability, 18, 2337–2366.
McKeague, I. W., Qian, M. (2015). An adaptive resampling test for detecting the presence of significant predictors. Journal of the American Statistical Association, 110, 1422–1433.
Portnoy, S. (2003). Censored regression quantiles. Journal of the American Statistical Association, 98, 1001–1012.
Shao, X., Zhang, J. (2014). Martingale difference correlation and its use in high-dimensional variable screening. Journal of the American Statistical Association, 109, 1302–1318.
Stute, W. (1997). Nonparametric model checks for regression. The Annals of Statistics, 25, 613–641.
Su, L., Zheng, X. (2017). A martingale-difference-divergence-based test for specification. Economics Letters, 156, 162–167.
van de Geer, S., Bühlmann, P., Ritov, Y., Dezeure, R. (2014). On asymptotically optimal confidence regions and tests for high-dimensional models. The Annals of Statistics, 42, 1166–1202.
Wang, H. J., Wang, L. (2009). Locally weighted censored quantile regression. Journal of the American Statistical Association, 104, 1117–1128.
Wang, H. J., McKeague, I. W., Qian, M. (2018). Testing for marginal linear effects in quantile regression. Journal of the Royal Statistical Society: Series B, 80, 433–452.
Xu, K., Chen, F. (2020). Martingale-difference-divergence-based tests for goodness-of-fit in quantile models. Journal of Statistical Planning and Inference, 207, 138–154.
Zheng, J. X. (1996). A consistent test of functional form via nonparametric estimation technique. Journal of Econometrics, 75, 263–289.
Zheng, J. X. (1998). A consistent nonparametric test of parametric regression models under conditional quantile restrictions. Econometric Theory, 14, 123–138.
Zhang, X., Yao, S., Shao, X. (2018). Conditional mean and quantile dependence testing in high dimension. The Annals of Statistics, 46, 219–246.
Acknowledgements
This work is supported by National Natural Science Foundation of China (12271005, 11901006), Young Scholars Program of Anhui Province and Natural Science Foundation of Anhui Province (1908085QA06).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Appendix: Necessary Lemmas
Appendix: Necessary Lemmas
We start with providing technical lemmas used repeatedly in the online Supplementary Material.
Lemma 3
(Shao and Zhang 2014, Proposition 2) Suppose the distribution function of Y satisfies Assumption (C3), then there exist \(\epsilon _0>0\) and \(c>0\) such that for any \(\epsilon \in (0,\epsilon _0)\),
Lemma 4
(Chernozhukov et al. 2015, Corollary 5.1) Let \(Z, Z_1,\ldots , Z_n\) be i.i.d. random variables taking values in a measurable space \((S, \mathcal {S}),\) Q denote a probability measure on the measurable space, \(\mathcal {F}\subset L^{2}(Q)\) be a pointwise measurable class of real-valued functions on S with measurable envelope F, and \(\mathcal {N}(\mathcal {F}, \Vert \cdot \Vert _{Q,2}, \delta )\) denote the \(\delta \)-covering number for \(\mathcal {F}\) with respect to the \(L^{2}(Q)\)-seminorm \(\Vert \cdot \Vert _{Q,2}.\) Suppose that \(\mathcal {F}\) is VC type, i.e., there exist constants \(A\ge e\) and \(V\ge 1\) such that \(\sup _{Q}\mathcal {N}(\mathcal {F}, \Vert \cdot \Vert _{Q,2}, \epsilon \Vert F\Vert _{Q,2})\le (A/\epsilon )^{V}\) where \(\sup _{Q}\) is taken over all finitely discrete distributions on S. Furthermore, suppose that \(0<E\{F^{2}(Z)\}<\infty \), and let \(\sigma ^{2}>0\) be any positive constant such that \(\sup _{f\in \mathcal {F}}E\{f^{2}(Z)\}\le \sigma ^{2}\le E\{F^{2}(Z)\}.\) Define \(B=E^{1/2}\{\max _{1\le i\le n}F^{2}(Z_i)\}\). Then
up to a universal constant.
Lemma 5
(Einmahl and Li 2008, Theorem 3.1) Let \(\textbf{z}_1,\ldots , \textbf{z}_n\) be independent centered random vectors in \(\mathbb {R}^p\) where \(p\ge 2\). Write \(\textbf{z}_i=(Z_{i1},\ldots ,Z_{ip})^{{\tiny {\textrm{T}}}}\) for \(i=1,\ldots ,n\). If \(E(\max _{1\le k\le p}\mid Z_{ik}\mid ^{r})<\infty \) for \(i=1,\cdots ,n,\) and some \(r>2\), then
About this article
Cite this article
Xu, K., An, N. A tuning-free efficient test for marginal linear effects in high-dimensional quantile regression. Ann Inst Stat Math 76, 93–110 (2024). https://doi.org/10.1007/s10463-023-00877-3
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10463-023-00877-3