The platform will undergo maintenance on Sep 14 at about 7:45 AM EST and will be unavailable for approximately 2 hours.
2014
DOI: 10.3406/estat.2014.10484
|View full text |Cite
|
Sign up to set email alerts
|

La régression quantile en pratique

Abstract: Les auteurs remercient les très nombreuses personnes qui ont contribué par leurs commentaires sur les premières versions successives de cet article à l'améliorer significativement, et plus particulièrement Cédric Afsa, Pascale Breuil, Elise Coudin, Jean-Michel Floch, Marine Guillerm, Jérôme Le, Simon Quantin et Olivier Sautory, ainsi que deux relecteurs anonymes de la revue. Ils restent seuls responsables des erreurs et approximations qui pourraient demeurer dans cet article. La méthode des moindres déviations… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
6
0
1

Year Published

2014
2014
2023
2023

Publication Types

Select...
7
1
1

Relationship

0
9

Authors

Journals

citations
Cited by 18 publications
(7 citation statements)
references
References 30 publications
0
6
0
1
Order By: Relevance
“…It is more robust than OLS because there is no need to do any distributional assumption about the error term, contrary to the normality assumption in the standard regression model. In clear, quantile regression is, by construction, particularly robust to extreme values (Haultfoeuille D' and Givord (), p. 7). The linear conditional quantile function can be estimated by solving the following minimization problem: trueγ̂nu(q)=argminγ(q)E{}ςq()YXγ(q)…”
Section: Discussionmentioning
confidence: 99%
“…It is more robust than OLS because there is no need to do any distributional assumption about the error term, contrary to the normality assumption in the standard regression model. In clear, quantile regression is, by construction, particularly robust to extreme values (Haultfoeuille D' and Givord (), p. 7). The linear conditional quantile function can be estimated by solving the following minimization problem: trueγ̂nu(q)=argminγ(q)E{}ςq()YXγ(q)…”
Section: Discussionmentioning
confidence: 99%
“…Moreover, in order to improve the robustness and to verify and confirm the veracity of the results obtained by the parametric model, it is necessary to estimate the econometric model according to the non-parametric model by using the quantile regression. Indeed, this latter is considered to be an integral part of the most appropriate statistical tools, which offers a richer and more flexible regression than classical linear regression (D'haultfoeuille & Givord, 2014). Therefore, by examining the results presented in the picture below, the overall estimation of the non-parametric model is similar as the parametric model considering that Considering the results of the quantile regression model, the independent variable corresponding to the tax burden is significant at the level of 1%, which is consistent with the results obtained by the Prais-Winsten regression model.…”
Section: Methodsmentioning
confidence: 99%
“…This questioning can't be enlightened under the hypothesis of linearity, is possible using the quantile regression methodology of Bassett (1978. According D'Holffoueille andGivord (2014), it's true because there is generally high heterogeneity in the data's from household surveys, as the data using in this study.…”
Section: Introductionmentioning
confidence: 92%