COMPSTAT 2016: Start Registration
View Submission - COMPSTAT
Title: Exponentiated-type distribution families with applications to scientific performance evaluation Authors:  Silvia Dedu - Bucharest University of Economic Studies (Romania) [presenting]
Vasile Preda - University of Bucharest (Romania)
Abstract: New exponentiated-type families of distributions are proposed and their properties are studied. The most relevant probabilistic descriptive measures are obtained, including moments, quantiles and concentration measures. Also, the distribution of order statistics is derived and several estimation methods are performed, based on moments, quantiles, least squares and maximum likelihood estimation. The relevance and effectiveness of the theoretical results are illustrated by solving several problems which consist in modeling and analyzing informetric distributions. The applications on real data sets containing impact factors from several relevant scientific fields prove that the use of these distribution families allows a higher degree of flexibility. Computational results are provided in order to highlight the advantages of our models.