2. 您的位置
3. 首页
4. 会议论文
5. Adaptive Quantile Regression with Precise Nonasymptotic Risk Bounds

Adaptive Quantile Regression with Precise Nonasymptotic Risk Bounds

来源 :The Third IMS-China International Conference on Statistics a | 被引量 : 0次 | 上传用户：seanchn

【摘 要】

：

Locally constant smoothing with adaptive choice of weights for nonparametric conditional quantile regression is considered.Some theoretical properties of the procedure are presented in the precise non

【作 者】

：

MaozaiTian Ngai Hang Chan 

【机 构】

：

Center for Applied Statistics, School of Statistics Remin University of China, Beijing, 100872, Chin

【出 处】

：

The Third IMS-China International Conference on Statistics a

【发表日期】

：

2011年3期

【关键词】

：

Adaptive smoothing Automatic Bandwidths Conditional quantiles Risk Bounds Robust 

下载到本地 , 更方便阅读

下载此文 赞助VIP

声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架

论文部分内容阅读

　　Locally constant smoothing with adaptive choice of weights for nonparametric conditional quantile regression is considered.Some theoretical properties of the procedure are presented in the precise nonasymptotic way.These results are technically more difficult and usually much more informative.

其他文献

Selective Reduced Rank Modeling for High-Dimensional Multivariate Data

Feature extraction corresponds to reduced rank modeling which has been widespread real-world applications.This talk investigates non-asymptotic performance of rank penalized estimators.The constructio

会议

Inference in Conditional Moment Restrictions Models in the Presence of Censoring on the Dependent Va

Much of the recent statistics, biostatistics and econometric literature investigate conditional moment restrictions (or estimating equations) models in the presence of censored observations.Parametric

会议

Extended BIC for linear regression models with diverging number of parameters and high or ultra-high

In many conventional scientific investigations with high or ultra-high dimensional feature spaces, the relevant features, though sparse, are large in number compared with classical statistical problem

会议

Sparse Matrix Graphical Models

We propose a novel sparse matrix graphical model for matrix-variate data.By penalizing the row and column precision matrices, our method gives sparse estimates of these matrices.The resulting estimate

会议

Extreme Volatility, Jump and Tail Risk: Dynamic Forecasting Stock Market Risk Based on Realized Vola

Using nonparametric estimation method, this paper studies the dynamic evolution features of jumps in volatility of domestic stock market.We build the heterogeneous autoregressive model of realized vol

会议

Realized VolatilityJumpTail RiskDynamic forecasting

An EM algorithm for penalized least squares

We propose an algorithm, called orthogonalizing EM (OEM), intended for fitting penalized regression models.It is motivated by the experimental design problem of fitting penalized regression with an or

会议

Smoothed log-concave maximum likelihood estimation with applications

We study the smoothed log-concave maximum likelihood estimator of a probability distribution in arbitrary dimensions.This is a fully automatic nonparametric density estimator, obtained as a canonical

会议

Efficient Estimation for Response Adaptive Designs

We discuss the estimation issues for response adaptive clinical trials.The asymptotic efficiency of maximum likelihood estimators of the parameters and the treatment effect are developed.We also explo

会议

Bootstrap testing multiple changes in persistence for a heavy-tailed sequence

There is a growing body of evidence showing that economic and financial time series display changes in persisteuce.This has been an issue of substantial empirical int crest, especially concerning infl

会议

Sequential Importance Sampling for Statistical Analysis of Network data

Random graphs with given vertex degrees have been widely used as a model for many real-world complex networks.We describe a sequential sampling method for sampling networks with a given degree sequenc

会议