Oil Prices: Breaks and Trends
The paper organized by the author also has an in-depth relationship with other literature works on the subject of Oil Breaks and Price Trends. Firstly, it has used the Kejriwal and Perron’s sequential procedure to determine the multiple structure changes in real oil prices. However, the author avoided the circular testing problem. In fact, the Kejriwal and Peron model allows the repeated use and selection of a number of breaks in the trend and in the level of a univariate time series without any prior knowledge as to whether the noise component is stationary or not. Thus, the author, in order to deal with the circular testing problem ,used the stability tests suggested by Vogelsang and Harvey. Apart from these academicians, Perron and Yabu developed an alternative test with higher power and less size distortions. Both, Perron and Yabu, used a super efficient estimate of the sum of the auto-regressive parameters and the break date is unknown. The model works by improving the finite sample properties where a bias corrected version of the OLS estimate is used. Thus, based on Perron and Yabu and the other testing strategies of Bai and Perron, Kejriwal and Perron during 2010 a sequential procedure to test the null hypothesis to test the null hypothesis of l changes against the alternative hypothesis of (l+1) changes.
In reference for the data used, the author conducted all the tests using log of real prices while the range of data was from January 1861 to August 2011. The data source was of monthly time series data of west Texas Intermediate and has been provided by the Global Financial Database. Since the original data was of nominal values, author deflated the series using Consumer Price Index issued by U.S. Bureau of Labor Statistics. Below are the summary of data and related results:
TABLE 1. BREAKDATES DETECTED BY THE KP (2010)
SEQUENTIAL PROCEDURE
Break Dates in Level
March 1930
February 1946
TABLE 2. UNIT ROOT TESTS
- H0: There is a unit root
- H0: The series is stationary (***) Rejected at 1% critical level (**) Rejected at 5% critical level
(*) Rejected at 10% critical level
TABLE 3. REGRESSION RESULTS
Variance Equation
The literature although had adequate accuracy, but some of the sentences were having grammar mistakes with inclusion of fragments in the sentence. Another issue of the literature was the language used by the author was not easy to understand although the range of data used was praiseworthy and reliable.
In order to make this literature better, we could aim for better explanation of economic reasons of such changes which may vary from exogenous events just as the Great Depression to political shakes as of energy crisis in 1970s.
In conclusion, a major section of the literature is attributed to work of Kejriwal and Perron’s sequential procedure in order to determine the structural changes in trend and level of real oil prices. Further, the outcome result was used to test the stationarity of real oil prices and to suggest the explanation for the structural changes found. The advantage of this method used over previous developed methods was that for this methodology, it was not necessary to know the integration level of the series in advance. In conclusion, the author using his methodology determined twelve structural changes for a given period from 1861 to 2011.
Works Cited
Noguera, Jose. Oil Prices: Breaks and Trends. Research. Chile: University of Santiago de Chile, n.d.
