Journal of Economic Modeling Research

Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

---

The characterization of memory effects in crude oil markets is an interesting issue that has attracted the attention of researchers from different disciplines, from econophysics to more classical economics. The importance of the problem relies in the fact that the departure from uncorrelated behavior would imply the presence of not-random effects which, in principle, can be exploited for arbitrage. This paper tries to contribute into the issue by estimating the memory effects by means of different parametric, semi-parametric, and non-parametric methods. In the other words, this paper provides analysis on the memory of the oil markets measured via the fractional integration parameter (d) by estimating it with various methods such as the MLE, NLS, GPH, Whittle, Lo, Hurst Exponent and Wavelet. To achieve this goal, we use the daily time series for WTI and Brent spot crude oil prices as well as 3-month futures, and further divide them into yearly subsections to obtain the historical series of memories. Results of the whittle and wavelet estimations, which are better suitted for this analysis, show no evidence of a long memory process. However, the oil price time series exhibits a nonstationary mean-reverting behaviour. Note that in this paper the behaviour of memory is our concern instead of the memory value itself. The results of memory changes trend shows that memory of international oil markets does not have an important trend change. In the other words, in our study period the efficiency of the market does not have an important decline or increase.

Keywords: Fractional Differencing, Oil Market Memory, ARFIMA Model, Time Series

Full-Text [PDF 985 kb]   (2123 Downloads)    

Add your comments about this article : Your username or Email:

---

---