2nd European International Conference on Industrial Engineering and Operations Management

Application of Geographically Weighted Regression Analysis on World’s Gasoline Prices Model

Tuti Purwaningsih
Publisher: IEOM Society International
0 Paper Citations
Track: Big Data and Analytics

This study aims to determine the effect of inflation, disaster risk, per capita GDP and global import export of world gasoline prices in 2017. The research is a kind of explanatory research with quantitative approach. The data used are nominal data with 5 independent variables namely inflation, GDP per capita, disaster risk, export and import in 2016 with the dependent variable of world gasoline price in 2017. The data used as many as 100 countries in the world are used as samples. With the method of spatial regression and weighted regression of geography the researcher wanted to know the influence of the five independent variables to the world gas price in 2017. The results of this study indicate that simultaneously (t test) using GWR analysis on GDP per capita variables and disaster risk significantly influence the world gasoline prices in 2017, while in the variable of inflation, exports, and imports do not have a significant effect on the world gas price in 2017. Based on the results of research, it is expected that decision makers and the authorities as a policy regulator can take the decision to control macroeconomic conditions in the country so that gasoline prices do not go affected and worsening economic conditions in a country. Similarly, existing disaster risk factors can be considered to make alternatives and other policies to save gasoline prices to keep them stable. Thus, the results of this study can also be used to gain attention in maintaining the stability of petrol and macroeconomic prices remain controlled.

Published in: 2nd European International Conference on Industrial Engineering and Operations Management, Paris, France

Publisher: IEOM Society International
Date of Conference: July 26-27, 2018

ISBN: 978-1-5323-5945-3
ISSN/E-ISSN: 2169-8767