The analysis of relationship between alternative financial instruments and stock prices with the nonlinear cointegration relationship: The Turkey sample

Linear time series models are frequently used in econometrics literature. However, linear methods may be insufficient in modeling economic and financial variables. Therefore, it is more accurate to prefer nonlinear methods in determining the asymmetric effects of these variables. The aim of this thesis is to analyze the asymmetric relationship between alternative financial instruments and stock prices of Turkey with the Nonlinear ARDL model approach by using quarterly data for the period 2003: Q1-2021Q3. In this respect, general information related to variables is discussed in the first section of the thesis. In the second section of the thesis, the econometric methodology used in the method is explained. Method and results take part in the third section of the thesis. In the study, both linear and nonlinear ARDL models are estimated for making a comparison. A long-term relationship has been determined between alternative financial instruments and stock prices in both models. While the BIST 100 index gives the same response in the short and the long run to positive and negative changes in gold, exchange rate, and interest rates, the nonlinear ARDL model responds differently to positive and negative changes. Additionally, according to the results of the Hatemi-J (2012) asymmetric causality test, an asymmetric causality relationship is found between the variables. A causal relationship is found from the positive components of the BIST100 index to the positive and negative components of the gold prices and found from the negative components of the BIST 100 index to the positive components of gold prices and the negative components of interest rates. While there is causality from the positive components of gold, exchange rate and interest rates to the negative components of the BIST100 index, a causality relationship is found from the negative components of gold, exchange rate, and interest rates to the positive components of the BIST100 index.