Regression Quantiles(分位数开山之作,Koenker, R. and G. Bassett ,Econometrica1978)
Regression Quantiles
Author(s): Roger Koenker and Gilbert Bassett, Jr.
Reviewed work(s):
Source: Econometrica, Vol. 46, No. 1 (Jan., 1978), pp. 33-50
Published by: The Econometric Society
Stable URL: http://www.jstor.org/stable/1913643 .
Accessed: 13/03/2012 ...
Regression Quantiles
Author(s): Roger Koenker and Gilbert Bassett, Jr.
Reviewed work(s):
Source: Econometrica, Vol. 46, No. 1 (Jan., 1978), pp. 33-50
Published by: The Econometric Society
Stable URL: http://www.jstor.org/stable/1913643 .
Accessed: 13/03/2012 13:03
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
The Econometric Society is collaborating with JSTOR to digitize, preserve and extend access to Econometrica.
http://www.jstor.org
Econometrica, Vol. 46, No. 1 (January, 1978)
REGRESSION QUANTILES'
BY ROGER KOENKER AND GILBERT BASSETT, JR.
A simple minimization problem yielding the ordinary sample quantiles in the location
model is shown to generalize naturally to the linear model generating a new class of
statistics we term "regression quantiles." The estimator which minimizes the sum of
absolute residuals is an important special case. Some equivariance properties and the joint
asymptotic distribution of regression quantiles are established. These results permit a
natural generalization to the linear model of certain well-known robust estimators of
location.
Estimators are suggested, which have comparable efficiency to least squares for
Gaussian linear models while substantially out-performing the least-squares estimator
over a wide class of non-Gaussian error distributions.
1. INTRODUCTION
IN STATISTICAL PARLANCE the term robustness has come to connote a certain
resilience of statistical procedures to deviations from the assumptions of
hypothetical models. The paradigm may be briefly stated as follows.2 The process
generating observed data is thought to be approximately described by an element
of some parametric class of models. The investigator seeks statistics, i.e., a
mapping from the sample space to a parameter space, whose distribution will be as
concentrated as possible near the true parameters-if the hypothesized model is
correct. If however, as seems almost certain, the parametric model is not quite
true, one would like to use estimators whose distributions were altered only
slightly if the distribution of the observations were close, in some reasonable
sense, to that of some member of the parametric class. In important special cases
this modest robustness requirement is not met by estimators in common use.3
We consider the familiar problem of estimating a vector of unknown (regres-
sion) parameters, 1B, from a sample of independent observations on random
variables Y,, Y2,. . ., YT, distributed according to
(1. 1) P(Yt
本文档为【Regression Quantiles(分位数开山之作,Koenker, R. and G. Bassett ,Econometrica1978)】,请使用软件OFFICE或WPS软件打开。作品中的文字与图均可以修改和编辑,
图片更改请在作品中右键图片并更换,文字修改请直接点击文字进行修改,也可以新增和删除文档中的内容。
该文档来自用户分享,如有侵权行为请发邮件ishare@vip.sina.com联系网站客服,我们会及时删除。
[版权声明] 本站所有资料为用户分享产生,若发现您的权利被侵害,请联系客服邮件isharekefu@iask.cn,我们尽快处理。
本作品所展示的图片、画像、字体、音乐的版权可能需版权方额外授权,请谨慎使用。
网站提供的党政主题相关内容(国旗、国徽、党徽..)目的在于配合国家政策宣传,仅限个人学习分享使用,禁止用于任何广告和商用目的。