Modelling U.S. gasoline demand: a structural time series analysis with asymmetric price responses

Zafer Dilaver, Lester Hunt

Research output: Contribution to journalArticlepeer-review

Abstract

This research aims to estimate a gasoline demand function for the U.S. using a stochastic exogenous trend model with asymmetric price responses. It is, as far as is known, the first attempt to model U.S. gasoline demand using this combined approach. The Structural Time Series Model is therefore employed for annual data over the period 1949-2019 allowing for both asymmetric price responses (for technical progress to affect demand endogenously) and an underlying energy demand trend for gasoline (for technical progress and other factors to affect demand exogenously in a linear or non-linear way). It is found that for U.S. per capita gasoline demand, the estimated long-run income elasticity is 0.41, the estimated long-run price-max elasticity is -0.31, the estimated long-run price-recovery elasticity is -0.15, and the estimated long-run price-cut elasticity is -0.14. In addition, the estimated underlying energy demand trend for U.S. per capita gasoline demand is non-linear with periods when it is increasing and periods when it is decreasing.
Original languageEnglish
Article number112386
Number of pages13
JournalEnergy Policy
Early online date16 Jul 2021
DOIs
Publication statusPublished - 1 Sep 2021

Keywords

  • U.S. gasoline demand
  • asymmetric price responses (APR)
  • underlying energy demand trend (UEDT)
  • structural time series model (STSM)
  • price and income elasticities

Fingerprint

Dive into the research topics of 'Modelling U.S. gasoline demand: a structural time series analysis with asymmetric price responses'. Together they form a unique fingerprint.

Cite this