eThesis - Kauppakorkeakoulun sähköiset gradut

» hse_ethesis_12839.pdf   koko: 3 MB (2964602)

Objective of the study

The objective of the study is to investigate how credit risk, liquidity and news about macroeconomic factors affect to Euribor basis swap spreads. Euribor basis swap spreads should trade in flat in order to no-arbitrage condition to hold. However, during the current financial crisis spreads have increased significantly. I will regard euro countries’ and Euribor panel banks’ credit default swap spreads as a credit risk component and the actions of the European Central Bank as a factors of liquidity component.

Data and methods

The study is empirical and will be based on linear regression and co-integration analysis. Data consist of time-series data from July 2008 to December 2011. I will first investigate relations between explanatory variables and 3 month versus 12 month Euribor basis swap spreads with 2 and 5 year maturities using descriptive statistics. After that, I will present proper empirical test results. In first phase, I will use unit root tests to see are time-series stationary. After they are stated to be stationary in differences and log-differences I will proceed to conduct short-run linear regression tests using ordinary least squares estimation. Finally, I will conduct Engle-Granger and Johansen co-integration tests in order to find out is there long run relationship between the explanatory variables and Euribor basis swap spreads.

Results

Results are in some sense twofold. Ordinary least squares provide rather different test results than co-integration tests for the short-run. In the long-run, I found only one robust co-integrating relation when applying both co integration methods. The relation was between Euribor basis swap spreads and Eurobond yield. In the shortrun I found five significant factors that could model the movements of Euribor basis swap spreads. The coefficients of determination were 30 per cent (2 year) and 25 per cent (5 year). Based on OLS results I could accept my initial hypothesis about significant components being credit risk, liquidity and news about macroeconomic variables. On the other hand based on co-integration tests I could accept liquidity component.