Thesis Type: Postgraduate
Institution Of The Thesis: Orta Doğu Teknik Üniversitesi, Faculty of Economic and Administrative Sciences, Department of Economics, Turkey
Approval Date: 2017
Student: ERDAL EROL
Supervisor: DİLEM YILDIRIM KASAP
Abstract:The major aim of this study is to find out the best models which identifies the determinants of food inflation in Turkey over the period 2003(1)-2013(12). To this end, a linear ARIMAX model is constructed by adopting an exhaustive search algorithm that is designed to select the model with most appropriate food inflation correlates. The best two models out of all candidate models having the minimum AIC and BIC values are evaluated in the estimation and prediction analyses. Roughly, our empirical results reveal that the main determinants of food inflation in Turkey are supply related factors such as producer prices of agriculture and food industry and price of water used in production. Moreover, exchange rate included in both models can be also classified as supply related variables due to its indirect effect on the input prices. To explore further the prediction performance of the selected models, a validation exercise is implemented for 2013-2014 and RMSE and MAE values of the models are compared. The results state that the best AIC model is as better due to the lower values of RMSE and MAE and the accuracy in the predicted values comparing to the best BIC model.