HAL will be down for maintenance from Friday, June 10 at 4pm through Monday, June 13 at 9am. More information
Skip to Main content Skip to Navigation

Econométrie des données imparfaites : méthodes et applications

Abstract : The Econometrics approach enables to formalize and stylize an economic phenomenon. Over time, econometricians face new problems related to changes in data. For example, the prevalence of massive data or new data protection reforms are changing the nature, the form or the availability of data. As a result, econometricians face new challenges and practical problems in dealing with data. The aim of this thesis is to propose and study methodologies adapted to imperfect data, more specifically, aggregated variables and limited dependent variables through two economic phenomena: first, inequality measures, and second, bodily injury claims. In each context, a set of relevant methods is proposed to deal with the available data and to explain each phenomenon. In the context of aggregated data, a study of compositional variables related to the aggregation of categorical data (Chapter I) and a method for estimating inequality measures from quantile data (Chapter II) are provided. In the context of censored data, a method for analyzing bodily injury claims amounts is proposed (Chapter III).
Document type :
Complete list of metadata

Contributor : Abes Star :  Contact
Submitted on : Friday, January 7, 2022 - 12:08:07 PM
Last modification on : Thursday, March 24, 2022 - 11:26:40 AM
Long-term archiving on: : Friday, April 8, 2022 - 6:50:25 PM


Version validated by the jury (STAR)


  • HAL Id : tel-03516637, version 1


Enora Belz. Econométrie des données imparfaites : méthodes et applications. Economies et finances. Université Rennes 1, 2021. Français. ⟨NNT : 2021REN1G002⟩. ⟨tel-03516637⟩



Record views


Files downloads