The demand for natural gas in the Northeastern United States

Highlights: • Natural gas demand is estimated in residential, commercial, and industrial sectors. • U.S. state-level data for natural gas (NG) use from 1997 to 2016 are utilized. • The AMG method shows better performance compared to other methods in the long run. • The NG demand is responsive to its own price and heating oil price in the long run. • The HDD has significant positive effects on NG demand in all three sectors. This paper examines the demand for natural gas in the residential, commercial, and industrial sectors of the Northeastern United States, comprising nine states and using annual state-level panel data over the period between 1997 and 2016. It applies panel unit root and cointegration tests, and then estimates the parameters using five alternative estimators: dynamic fixed effects (DFE), mean group (MG), pooled mean group (PMG), common correlated effect mean group (CCEMG), and augmented mean group (AMG). The panel unit root and cointegration tests show that the series are I (1), and cointegrated. The estimated results show that the long run own price elasticities for natural gas in residential, commercial, and industrial sectors are −0.14, −0.29, and −0.28, respectively. The cross price elasticities of fuel oil for natural gas demand in residential, commercial, and industrial sectors are 0.19, 0.52, and 0.24, respectively. The long run natural gas demand is not affected by income in all three sectors. The heating degree days (HDD) have significant positive effects on demand for natural gas in all three sectors.