Predicting World Energy Consumption: Comparison of ANN and Regression Analysis

World energy consumption is responsible for the growth of the economy. Having it in abundant will do more to the continuous global growth. Fundamental factors to how energy is consumed are population and gross domestic product (GDP). This study focusses on predicting the global energy consumption fr...

Full description

Saved in:
Bibliographic Details
Main Authors: Olanrewaju, O. A., Mbohwa, C.
Format: Conference Proceeding
Language:English
Subjects:
ANN
Artificial neural networks
Economic indicators
Energy consumption
GDP
Neurons
Population
Regression analysis
Sociology
World Energy Consumption
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:World energy consumption is responsible for the growth of the economy. Having it in abundant will do more to the continuous global growth. Fundamental factors to how energy is consumed are population and gross domestic product (GDP). This study focusses on predicting the global energy consumption from 1995 to 2009 using the fundamental factors as inputs. Statistical and evolutionary algorithm in the form of regression analysis and artificial neural network (ANN) were compared in their prediction performance. Both techniques performed brilliantly as indicated by the coefficient of correlation and visual inspection, however, ANN performed better. Analyzing the factors through the connection weights of ANN reported population to be the significant factor contributing more to how energy is consumed globally. It is important to have policies that can influence population positively in order to have abundant supply of energy whenever there is demand for it.
ISSN:2156-017X
DOI:10.1109/IEDM19572.2017.9110625