SOVEREIGN CREDIT DEFAULT SWAP (CDS) SPREADS CHANGES IN VARIOUS ECONOMIC CONJUNCTURES: EVIDENCE FROM TURKEY BY MACHINE LEARNING ALGORITHMS

The study aims to define the sources of Turkey’s sovereign CDS spread changes develop policies that stabilize spreads since they have a volatile and increasing trend, especially in last two years. In this context, monthly data 13 factors related international, macroeconomic, market between 2011/1 2019/12 are used by dividing dataset into three periods as full period (2011-2019), stability (2011-2017), macroeconomic turbulent (2018-2019) performing 4 different machine learning algorithms. empirical results prove (i) Treasury bond interest rate should be lower than 8% gold prices TL 5.500 low-level spreads; (ii) NPL volume has no significant effect on any examined; (iii) significance vary over periods.