Stochastic level shifts and outliers and the dynamics of oil price movements

Abstract

Oil prices clearly play an important role in the macroeconomy. The dynamics of oil prices have, however, been difficult to pin down because of the frequent occurrence of large shocks. In this paper, we propose a time series model with heavy-tailed disturbances to analyze the dynamics of the oil price. The model has the form of a generalized local linear trend, and we show that it successfully captures outliers and level shifts as empirical regularities in the oil price, including known historical price shocks. Further, the results of a forecast exercise are given, and we study extensions that examine the effect of the GDP cycle on the oil price.