Skip navigation
Please use this identifier to cite or link to this item: http://arks.princeton.edu/ark:/88435/dsp01q237hv31f
Title: Vector Autoregression Analysis of the Sovereign Credit Default Swap Returns of Eurozone Economies
Authors: Wu, Liukun
Advisors: Fan, Jianqing
Department: Operations Research and Financial Engineering
Class Year: 2015
Abstract: This paper investigates the interactions among the sovereign CDS daily log returns of ve major European Monetary Union countries (Germany, France, Italy, Portugal, and Spain) involved in the Eurozone debt crisis, from the period 2010 to 2014. We form a vector autoregression model (VAR) with the ve countries' CDS daily log returns as endogenous variables, and exogenous variables such as domestic factors (country's equity market return index), global risk factors (the VIX Index), and global macroeconomic factors (US Treasury daily yields) to account for the correlation among the sovereign CDS price series. We conditionally forecast the log returns three- month ahead using month by month rolling for maturities of 1, 5, and 10 years. We then calculate the root-mean-squared error (RSME) and compare that with the direct least squares and ARMA model forecast. We nd that the VAR model gives stable RMSE values of 0.03 on average for each sovereign for the 5 and 10 year maturities, 0.06 for the 1 year maturity, and performs slightly better than the direct least squares. The RMSE values from the ARMA model are similar to that of VAR model, but the VAR model is able to describe more characteristics in the return series and matches the volatility pattern better.
Extent: 100 pages
URI: http://arks.princeton.edu/ark:/88435/dsp01q237hv31f
Type of Material: Princeton University Senior Theses
Language: en_US
Appears in Collections:Operations Research and Financial Engineering, 2000-2016

Files in This Item:
File SizeFormat 
PUTheses2015-Wu_Liukun.pdf1.25 MBAdobe PDF    Request a copy


Items in Dataspace are protected by copyright, with all rights reserved, unless otherwise indicated.