冯兴东教授带领团队在分位数回归的理论和方法领域开展高水平研究
分位数回归已经在诸如金融、经济学、气象学以及生物学等多个研究领域成为传统均值回归有效的替代性模型。这是由于分位数回归具备三个重要特征：（i）分位数回归研究响应变量在不同分位数水平上与解释变量之间的关系，从而可以更为全面地反映出数据蕴含的重要信息；(ii)分位数回归对应的损失函数拥有有界的一阶导数，因此从理论上具备了稳健性，可以较好处理响应变量中出现的异常值；(iii)分位数回归通常不对响应变量加以分布假设，允许模型误差具备异质性。如果我们仅仅研究指标的均值情况，那么大量复杂的实际现象将无法得到有效分析。冯兴东教授带领团队在分位数回归领域长期耕作，取得一系列有影响力的理论方法成果[1-13]。例如，由于分位数回归模型中的参数估计通常涉及到对于密度函数的估计，而密度函数的估计通常不太稳定，因此再取样方法就成为了一种比较好的备选方案。冯兴东教授提出了一种针对分位数回归的新的再取样方法，该方法可以容忍分位数回归模型中模型误差的异方差性。该方法的提出迅速受到学术界的关注，在第一时间被整合入分位数回归模型创始人R. Koenker开发和维护的R语言软件包quantreg之中。该文章自2011年发表在Biometrika之后获得不少应用领域的引用[3]。
参考文献：
[1] Dong, C.#, Li, G. and Feng, X.* (2019). Lack-of-fit tests for quantile regression models. Journal of the Royal Statistical Society B 81, 629-648.
[2] Dong, C.#, Ma, S., Zhu, L., Feng, X.*(2021). Estimation and inference for non-crossing multiple-index quantile regression(in Chinese). SCIENTIA SINICA Mathematica 51, 631-658.
[3] Feng, X., He, X. and Hu, J.* (2011). Wild bootstrap for quantile regression. Biometrika 98, 995-999.
[4] Feng, X., He, X.*, Jiao, Y., Kang, L.*, and Wang, C.# (2024). Deep nonparametric quantile regression under covariate shift. Journal of Machine Learning Research 25(385), 1-50.
[5] Feng, X., Li, W.#, and Zhu, Q.* (2023). Spatial-temporal model with heterogeneous random effects. Statistica Sinica 33, 2613-2641.
[6] Feng, X.*, Li, W.#, and Zhu, Q. (2024). Estimation and bootstrapping under spatiotemporal models with unobserved heterogeneity. Journal of Econometrics 238, 105559.
[7] Feng, X.*, Liu, Q.#, and Wang, C.# (2023). A lack-of-fit test for quantile regression process models. Statistics and Probability Letters 192, 109680.
[8] Feng, X. and Zhu, L.* (2016). Estimation and testing of varying coefficients in quantile regression. Journal of the American Statistical Association 111, 266-274.
[9] Wang, C.#, Li, T., Zhang, X.#, Feng, X., and He, X.* (2024). Communication-efficient nonparametric quantile regression via random features. Journal of Computational and Graphical Statistics 33, 1175-1184.
[10] Wang, H. and Feng, X.* (2012). Multiple imputation for M regression with censored covariates. Journal of the American Statistical Association 107, 194-204.
[11] Wang, H., Feng, X.* and Dong, C.# (2019). Copula-based quantile regression for longitudinal data. Statistica Sinica 29, 245-264.
[12] Yi, Y., Feng, X., and Huang, Z.* (2014). Estimation of extreme value-at-risk: an EVT approach for quantile GARCH model. Economics Letters 124, 378-381.
[13] Yu, A.#, Zhong, Y., Feng, X.*, and Wei, Y. (2023). Quantile regression for nonignorable missing data with its application of analyzing electronic medical records. Biometrics 79, 2036-2049.
