Analysis of gas prices for Turkey from 2003-2011

---

Tezin Türü: Yüksek Lisans

Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Fen Edebiyat Fakültesi, İstatistik Bölümü, Türkiye

Tezin Onay Tarihi: 2012

Öğrenci: WİLBERFORCE KİRİBAKİ

Danışman: CEYLAN YOZGATLIGİL

Özet:

This study aimed to construct a forecasting model for gas prices in Turkey using Univariate time series analysis. The best model was developed after assessing the forecasting performances for both Seasonal Autoregressive Integrated Moving Average (SARIMA) model and Exponential Smoothing (ES) model. Firstly, we fitted different combinations of both ARIMA and SARIMA models (from which the best model was chosen) by using the monthly oil prices from January 2003 to December 2011. The ES model was automatically fitted next for forecasting performance comparison purposes. We then extracted the forecasted monthly values for 2012 for both models and compared their forecast performances. The ARIMA (1,1,0) model gave the best fit for the gas price series for Turkey. The results depicted that the ARIMA model forecasts work effectively and reliably, and is a useful tool for forecasting future Turkish gas prices. It can be used by governments, investors and other gas users to predict and address negative impacts that gas shocks creates.