Nonlinear Quantile Regression for Longitudinal Data

Author: Andreas Karlsson; Uppsala Universitet.; [2005]

Keywords: NATURVETENSKAP; NATURAL SCIENCES; NATURVETENSKAP; NATURAL SCIENCES; bootstrap; dependent errors; median regression; Monte Carlo simulation; panel data; repeated measures; SOCIAL SCIENCES Statistics; computer and systems science Statistics; SAMHÄLLSVETENSKAP Statistik; data- och systemvetenskap Statistik; SOCIAL SCIENCES Statistics; computer and systems science Statistics Biostatistics; SAMHÄLLSVETENSKAP Statistik; data- och systemvetenskap Statistik Biostatistik;

Abstract: The overall objective of the two papers in this thesis is to examine the properties of the weighted nonlinear quantile regression estimator for the analysis of longitudinal data. To this end, the question of which weights to be used, the bias of the estimator and the possibility to calculate confidence intervals has to be examined. The focus is on small samples, since this is the most common case for real world applications in biostatistics.

  This dissertation MIGHT be available in PDF-format. Check this page to see if it is available for download.