Essays in volatility and risk modelling in interest rate swaps
2017-03-22T01:44:46Z (GMT) by
This thesis focuses on the linkages between volatility of interest rate swaps (hereafter, IRS) and macroeconomic risk, cross-border linkages of swap markets from two-factor volatility models, and the influence of three additional risk factors on swap spreads. In order to investigate these, the thesis presents three empirical research essays that all revolve around a common theme: volatility and risk modelling in interest rate swaps. First research essay, presented in Chapter 3, explores whether and how the volatility of swap yield curves is related to macroeconomic risk. The methodology in this essay is based on a recent Spline-GARCH model, multivariate regression, principal component analysis and Granger causality. The empirical analysis is conducted on a sample of daily data for the period between 1987 and 2010 from three major swap markets namely, Japan, the UK and the US. The empirical analysis reveals two important findings. First, using “low-frequency” volatility extracted from aggregate volatility shocks of the three swap markets the analysis suggests that this low-frequency IRS volatility has strong and (mostly) positive association with most of the macroeconomic risk proxies. This relationship between the macroeconomic risks and IRS volatility varies slightly across the different swap maturities but is robust to alternative volatility specifications, namely C-GARCH model and model-free realized volatility. This finding is fairly consistent with the argument that the greater the macroeconomic risk the greater is the use of derivative instruments to hedge or speculate. Second, to explore the dynamic interaction including lead-lag relationship, the study finds that it is the low-frequency (IRS) volatility that Granger causes most of the macroeconomic risk proxies. This finding is, nonetheless, consistent with the argument that, as forward looking instrument, IRS has predictive power to forecast the changes in macroeconomic risk. Motivated by these findings, an empirical analysis is done on reverse regression in which macroeconomic risk proxies and their principal components are regressed on low-frequency volatility of swaps. The findings are encouraging for those who would like to use swaps in predicting macroeconomic risk. Second research essay, presented in Chapter 4, explores whether the observed relationship between macroeconomic risk proxies and volatility of swap market can be extended to investigate the cross-border linkages of swap markets. The mixed and inconclusive evidence on volatility transmission and swap market integration motivated this essay to investigate this issue from different approach. In particular, using the decomposed volatilities (long-term and short-term), this essay examines the financial integration and volatility linkages of three major swap markets, namely Japan, the UK and the US. To facilitate empirical investigation, a step-by-step approach is proposed in measuring volatility transmission and financial linkages including dynamic correlations, contagion and causality of volatility components. These findings have important implications for portfolio risk diversifications in swaps. Third research essay, presented in Chapter 5, exploits the puzzle with regard to determinants and components of swap spreads. This essay argues that in addition to default risk and liquidity risk, three risk factors namely, business cycle risk, market skewness risk and correlation risk contain significant information in determining the swap spreads. Using the GMM approach, this essay provides empirical support of risk premia related to these three risk factors in addition to default and liquidity risk premia. The results are robust to sub-sample analysis.