Modelling the natural gas consumption in a changing environment

Article type: Research Article

Authors: Batzias, F.A.^{; *} | Nikolaou, N.P. | Kakos, A.S.

Affiliations: Department of Industrial Management and Technology, University of Piraeus, 80 Karaoli & Dimitriou St., 185 34 Piraeus, Greece

Correspondence: [*] Corresponding author. E-mail: fbatzi@unipi.gr.

Abstract: A composite function was used successfully for modelling the Natural Gas (NG) consumption in 16 European energy markets. Background of the model is a logistic function where the upper limit is also a logistic function of time, with secondary parameters determined either endogenously together with the rest primary parameters or exogenously in a sample space of the energy market. Fitting of this ‘double logistic’ dynamic model to NG consumption data of the period 1980–2000 gave better Standard Errors of Estimate (SEEs) for ten energy markets in comparison with the linear, the exponential/asymptotic and the static logistic model. Supplementary results obtained by statistical analysis of answers selected by circulating a questionnaire in the wider area of Attica in Greece, led to the conclusion that income/welfare, residential place and information play an important role as regards the intention of inhabitants to adopt the NG alternative. These results can be used for policy making as regards incentives offered to potential consumers (that constitute a dynamic market share rather than a static one) to accelerate the adoption of NG technology.

Keywords: Natural Gas, forecasting, logistic function, technology dissemination, nonlinear regression, model selection, model discrimination

DOI: 10.3233/JCM-2006-65-607

Journal: Journal of Computational Methods in Sciences and Engineering, vol. 6, no. 5-6, pp. 339-352, 2006