Ali Gençay Özbekler, M.Sc.

Department of Financial Mathematics

August 2019

*Supervisor*: Ömür Uğur (Institute of Applied Mathematics, Middle East Technical University, Ankara) *Co-Supervisor*: İrem Talaslı (Central Bank of the Republic of Turkey)

### Abstract

We follow novel procedure of the work by Andersen T. G. and Benzoni L. (2010) to assess presence of unspanned stochastic volatility (USV) phenomenon in the Turkish lira interest rate swap (IRS) market. The estimations reveal that IRS yield curve dynamics fail to span volatility in IRS market and thus volatility risk cannot be hedged using only IRS instruments. The major implication of USV is then used to investigate the systemic volatility in domestic markets. In this scope, we employ USV condition as a speciﬁcation for afﬁne term structure (AFTS) models. Comparing AFTS models with stochastic and constant volatility, we ﬁnd that three-factor constant volatility model provides more robust estimation results in terms of both volatility and yield ﬁtting.

*Keywords*: Afﬁne Term Structure Models, Term Premia, Stochastic Volatility, Systemic Volatility, Spanning Hypothesis