Περίληψη: | The market of crude oil and derivative products that derive from it, constitute a
high percentage of energy used by people and businesses in economic activity. So
we expect a positive correlation between the price of Brent crude oil and the retail
gasoline price.
In this dissertation we examine the causal relationship between the retail price
of gasoline and the price of Brent crude oil and investigate possible price asymmetries in terms of convergence in the long-term balance that connects the two
variables. We also examined the causal relationship between the retail price and
gasoline price for the Greek oil market for the period from January 2005 to March
2020 with frequency, weekly observations.
For econometric specification we used the symmetric and asymmetric error
correction model. In addition, we present the results of the fully modified ordinary least squares estimator (FMOLS) and the dynamic ordinary least squares
estimator (DOLS) in order to compare the stability and the robustness of the
results.
The contribution of the article is based on the detailed description of the error
correction models of both symmetric and asymmetric. Also the contribution lies
in the empirical analysis and the finding of asymmetry in the price of gasoline in
the Greek oil market and the stability of the results from the two methods used
in the valuation process.
From empirical analysis, we show that time series are cointegrated and that
there is a long-term relationship between variables. The main conclusions of the
symmetric model showed that the retail price performance of gasoline depends on
the performance of gasoline with five lags and the performance of crude oil with
five lags. Also from the asymmetric error correction model we observed that the
rate of convergence is greater when we are below the long-term equilibrium than
when we are above the long-term equilibrium. The results shall be verified both
FMOLS and DOLS which take into account the asymptotic bias.
|