A conditionally heteroskedastic model with time-varying coefficients for daily gas spot prices

Nazim Regnard, Jean Michel Zakoïan

Research output: Contribution to journalArticlepeer-review

Abstract

This paper examines the relationship between gas spot prices at the Zeebrugge market, one-month ahead Brent prices and temperatures over the period 2000-2005. A cointegration analysis is carried out and it is discovered that a cointegration relationship exists between the three series. To take into account the influence of temperature on the gas volatility, a GARCH(1,1) model with temperature-dependent coefficients is considered. Stability and estimation properties are discussed. An empirical finding is the existence of distinct volatility regimes for the volatility of gas prices, depending on the temperature level.

Original languageEnglish
Pages (from-to)1240-1251
Number of pages12
JournalEnergy Economics
Volume33
Issue number6
DOIs
Publication statusPublished - 1 Nov 2011
Externally publishedYes

Keywords

  • GARCH
  • Gas prices
  • Nonstationary models
  • Periodic models
  • Quasi-maximum likelihood estimation
  • Time-varying coefficients

Fingerprint

Dive into the research topics of 'A conditionally heteroskedastic model with time-varying coefficients for daily gas spot prices'. Together they form a unique fingerprint.

Cite this