Session

Computer Science and Information Systems

Description

In the paper “The analysis of compositional data, a general overview and an application with GDP data for Albanian economy in R software" we have studied the principle rules of the compositional data analysis. We have listed some of the fields where we can find and we can apply compositional data analysis. Furthermore there have been treated the main problems that a user will have during the work with coda data. After problems there are a lot of ways and methods in order to avoid those problems and some transformations that really help the coda work. The most important part of this work will be considered the application that we have separated it into two parts. We have chosen the GDP data, because we can consider them as compositional data. From every model we have concluded some important results and we have compared some parameters and results too. As a conclusion we have introduced the idea for a further work.

Keywords:

compositional data analysis, model, predictions, GDP

Session Chair

Kozeta Sevrani

Session Co-Chair

Galia Marinova

Proceedings Editor

Edmond Hajrizi

ISBN

978-9951-437-31-8

First Page

71

Last Page

83

Location

Durres, Albania

Start Date

8-11-2014 9:00 AM

End Date

8-11-2014 9:15 AM

DOI

10.33107/ubt-ic.2014.60

Share

COinS
 
Nov 8th, 9:00 AM Nov 8th, 9:15 AM

The analysis of composıtıonal data, a general overvıew and an applicatıon with GDP data for Albanıan economy in R software

Durres, Albania

In the paper “The analysis of compositional data, a general overview and an application with GDP data for Albanian economy in R software" we have studied the principle rules of the compositional data analysis. We have listed some of the fields where we can find and we can apply compositional data analysis. Furthermore there have been treated the main problems that a user will have during the work with coda data. After problems there are a lot of ways and methods in order to avoid those problems and some transformations that really help the coda work. The most important part of this work will be considered the application that we have separated it into two parts. We have chosen the GDP data, because we can consider them as compositional data. From every model we have concluded some important results and we have compared some parameters and results too. As a conclusion we have introduced the idea for a further work.