The lasso for high dimensional regression with a possible change point.

The lasso for high dimensional regression with a possible change point.
复制标题

DOI:
10.1111/rssb.12108
复制
发表时间:
2016-01
期刊:
Journal of the Royal Statistical Society. Series B, Statistical methodology
影响因子:
--
通讯作者:
Shin Y
Shin Y
中科院分区:
其他
文献类型:
--
作者:
Lee S;Seo MH;Shin Y

文献摘要

参考文献

被引文献

相似文献

本文考虑了一个高维回归模型,由于协变量阈值的存在,模型可能存在一个变点,并给出了回归系数和阈值参数的Lasso估计。我们的Lasso估计不仅选择协变量,而且选择线性和阈值回归模型之间的模型。在稀疏性假设下,我们得到了回归系数的预测风险和估计损失的非渐近预言不等式。由于拉索估计同时选择变量,我们表明,甲骨文不等式可以建立没有预先测试的阈值效应的存在。此外,我们建立了条件,在该条件下,当回归变量的数量远大于样本量n时,未知阈值参数的估计误差可以由几乎均匀的因子限制。我们说明了我们提出的估计方法的实用性,通过蒙特卡罗模拟和应用程序的真实的数据。
We consider a high dimensional regression model with a possible change point due to a covariate threshold and develop the lasso estimator of regression coefficients as well as the threshold parameter. Our lasso estimator not only selects covariates but also selects a model between linear and threshold regression models. Under a sparsity assumption, we derive non‐asymptotic oracle inequalities for both the prediction risk and the ‐estimation loss for regression coefficients. Since the lasso estimator selects variables simultaneously, we show that oracle inequalities can be established without pretesting the existence of the threshold effect. Furthermore, we establish conditions under which the estimation error of the unknown threshold parameter can be bounded by a factor that is nearly even when the number of regressors can be much larger than the sample size n. We illustrate the usefulness of our proposed estimation method via Monte Carlo simulations and an application to real data.
DOI: 10.1214/009053604000000256
发表时间: 2004-06-01
影响因子: 4.5
作者:
Fan, JQ;Peng, H
通讯作者: Peng, H
DOI: 10.1214/11-aos911
发表时间: 2011
影响因子: 4.5
作者:
Bradic J;Fan J;Jiang J
通讯作者: Jiang J
DOI: 10.1214/07-aos582
发表时间: 2009-02-01
影响因子: 4.5
作者:
Meinshausen, Nicolai;Yu, Bin
通讯作者: Yu, Bin
DOI: 10.1214/10-aos827
发表时间: 2011-02-01
影响因子: 4.5
作者:
Belloni, Alexandre;Chernozhukov, Victor
通讯作者: Chernozhukov, Victor
DOI: 10.1162/qjec.2008.123.1.177
发表时间: 2008-02-01
影响因子: 13.7
作者:
Card, David;Mas, Alexandre;Rothstein, Jesse
通讯作者: Rothstein, Jesse