Akademik Veri Yönetim Sistemi
The 7th International Conference on Computational Mathematics and Engineering Sciences / 20-21 May. 2023, Elazığ – Türkiye, Elazığ, Türkiye, 20 - 22 Mayıs 2023, ss.111, (Özet Bildiri)
Crude oil is known as the lifeblood of an economy, therefore, fluctuations in oil prices
have significantly affected many countries around the world, especially in the last few years,
due to various global effects. For this reason, in this study, the WTI Cushing, Oklahoma FOB
crude oil closing data between 01.03.2019 and 13.03.2023 are considered with SDE modeling
as they follow a trajectory in the form of the standard Wiener process. First, the WTI dataset is
modeled with the GBM SDE and the CIR SDE equations which are widely used in finance,
without considering the change point (CP) estimation. Then, the GBM SDE and the CIR SDE
models were reconstructed considering the CP estimation. Finally, a Hybrid SDE model is
proposed, which is compatible with WTI data, again considering the CPs. To obtain a hybrid
model, the GBM SDE and the CIR SDE models were used. The parameters in the model were
estimated by the quasi-maximum likelihood estimation method and the approximate solution
of each established SDE model was obtained with the Euler-Maruyama numerical solution. To
determine the best model for the data set among the established models, AIC, BIC, RMSE, and
MAPE criteria were used. Accordingly, the most suitable model for the data set according to
the RMSE and the MAPE criterion is the Hybrid SDE model. Unlike the other models, this
model explained the sudden ups and downs in the data set better.