Go to content
UR Home

Spline-based model specification and prediction for least squares and quantile regression

Kagerer, Kathrin (2014) Spline-based model specification and prediction for least squares and quantile regression. PhD, Universität Regensburg.

[img]
Preview
License: Publishing license for publications excluding print on demand
PDF
Download (6MB)
Date of publication of this fulltext: 22 May 2014 14:19

Abstract (English)

This dissertation focuses on non- and semiparametric specification of regression models for the conditional expectation and for conditional quantiles. For modeling the nonparametric component, essentially B-splines are applied. Different aspects of estimation and/or prediction are emphasized in the chapters and are applied in empirical as well as in simulated analyses.

Translation of the abstract (German)

Die Dissertation beschäftigt sich mit der nicht- bzw. semiparametrischen Spezifikation von Regressionsmodellen für den bedingten Erwartungswert und für bedingte Quantile. Dabei werden vor allem B-Splines verwendet, um die nicht-parametrische Komponente zu modellieren. In den einzelnen Kapiteln werden unterschiedliche Aspekte der Schätzung und/oder Prognose hervorgehoben und das Vorgehen in empirischen sowie in simulierten Analysen angewandt.


Export bibliographical data



Item type:Thesis of the University of Regensburg (PhD)
Date:22 May 2014
Referee:Prof. Dr. Rolf Tschernig
Date of exam:16 March 2012
Institutions:Business, Economics and Information Systems > Institut für Volkswirtschaftslehre und Ökonometrie > Lehrstuhl für Ökonometrie (Prof. Dr. Rolf Tschernig)
Keywords:regression, splines, prediction, quantile regression, model specification, hat matrix
Dewey Decimal Classification:300 Social sciences > 310 General statistics
Status:Published
Refereed:Unknown
Created at the University of Regensburg:Yes
Item ID:29858
Owner only: item control page

Downloads

Downloads per month over past year

  1. Homepage UR

University Library

Publication Server

Contact:

Publishing: oa@ur.de

Dissertations: dissertationen@ur.de

Research data: daten@ur.de

Contact persons