Estimation of demand function for natural gas in Iran: Evidences based on smooth transition regression models

This paper attempts to study the demand function of natural gas in Iran using smooth transition regression model (STR). To this end, gross domestic productions (GDP), real price of natural gas and temperature have been employed as variables explaining the natural gas demand from 1971 to 2009. The results indicate that natural gas demand follows an LSTR1 nonlinear two-regime model if the real price of natural gas is considered as the transition variable. Estimation results also indicate that the slope parameter approximately equals the high value of 10 and the threshold value is 31.82 Rials1 per one cubic meter consumption of natural gas. The results strongly suggest that natural gas demand follows the gross domestic production during the last years, but temperature has no significant impact on natural gas demand in Iran.