Price elasticity of residental natural gas demand in Turkey: A panel data application

In this study, aim is to both analyze the factors that affect the natural gas demand in households in Turkey and the price elasticity for the natural gas demand. In the study, cities that are eligible for analysis are included with the countrywide analysis for short and long period of time. Considering the time period between June 2015 and October 2019, Turkey's 54 cities' monthly data are obtained and according to these data sets, appropriate panel data analysis methods are used. The cities that are being included in this thesis study, create the natural gas demand and number of subscriber count with a minumum level of 93%. In the constructed model, households' natural gas demand, natural gas price, number of subscribers and temperature differences are introduced as variables. First, cross section dependency test, then CIPS test, which is one of the second generation unit root tests, was applied. In the short-term analysis of the unit effect and heterogeneity test results, the Amplified Mean Group (AMG) estimator was used. In order to examine the relation between the variables for long period of time cointegration test was used. Regarding the long term elasticity coefficents examination, DOLSMG model was used. In the short and long term, all findings about the elasticity related to countrywide and all cities included to analysis. Cointegration test is performed to examine the long-term relation of parameters/variables. The model of DOLSMG is taken into consideration with the determination of elasticity coefficients. The short and long-term findings of both the analysed cities and Turkey in general are shared in the study. For the short period of time, findings in Turkey suggest that, natural gas demand has inelastic relation with the price considering houses. Moving on to the different property, with the increase in the number of subscription, natural gas demand is observed as positive while increase in temperature is observed as negative. For the long period of time, in houses, relationship between the natural gas demand and variables is defined as cointegrated. Similar to the findings in the short period of time, natural gas demand is found as inelastic with the price in the long period of time. Natural gas demand is affected positive while increase in temperature is negative with the increase in the number of subscriptions. All the findings in the thesis are supported by the studies done earlier on the literature and eligible to the expectations.