MODELLING THE BIVARIATE DEPENDENCE STRUCTURE OF EXCHANGE RATES BEFORE AND AFTER THE INTRODUCTION OF THE EURO: A SEMI-PARAMETRIC APPROACH

被引:39
作者
Boero, Gianna [2 ]
Silvapulle, Param [1 ]
Tursunalieva, Ainura [1 ]
机构
[1] Monash Univ, Dept Econometr & Business Stat, Clayton, Vic 3800, Australia
[2] Univ Warwick, Dept Econ, Coventry CV4 7AL, W Midlands, England
关键词
Copulas; non-parametric plots; semi-parametric methods; tail dependence; ERROR DISTRIBUTION; MULTIVARIATE;
D O I
10.1002/ijfe.434
中图分类号
F8 [财政、金融];
学科分类号
0202 ;
摘要
This paper investigates the bivariate dependence structure for three pairs of exchange rates measured against the US dollar: Euro and Japanese yen (JY), Euro and GBP, Euro and Swiss franc (CHF), in the pre-, post-Euro and the transition periods with the sample period ranging from January 1994 to November 2007. The Deutsche mark (DM) is used for the pre-Euro period. The novelty of the paper is that it employs non-parametric plots, which were derived based on the concept of copula, and a robust semi-parametric method to estimate copula models. The results indicate the changes in the dependence structure from the pre-Euro to the post-Euro period for the pairs DM (Euro)-JY, and DM (Euro)-GBP, with major changes occurring during the initial years of the launch of the Euro. For these two pairs of exchange rates, the model captures asymmetric tail dependence, implying different degrees of co-movements during appreciations and depreciations against the USD. The dependence between the Euro and the CHF remains unchanged, both in strength and structure, over the whole sample period, reflecting a marked tendency of the CHF to follow the fluctuations of the Euro against USD. There has been a tendency for the dependence between Euro and GBP to increase across the whole sample period. The results may be of interest for central banks, international trade, international portfolio diversification and currency risk management. Copyright (C) 2010 John Wiley & Sons, Ltd.
引用
收藏
页码:357 / 374
页数:18
相关论文
共 33 条
[21]   The Copula-GARCH model of conditional dependencies: An international stock market application [J].
Jondeau, Eric ;
Rockinger, Michael .
JOURNAL OF INTERNATIONAL MONEY AND FINANCE, 2006, 25 (05) :827-853
[22]   A new measure of rank correlation [J].
Kendall, MG .
BIOMETRIKA, 1938, 30 :81-93
[23]   Estimating the error distribution in multivariate heteroscedastic time-series models [J].
Kim, Gunky ;
Silvapulle, Mervyn J. ;
Silvapulle, Paramsothy .
JOURNAL OF STATISTICAL PLANNING AND INFERENCE, 2008, 138 (05) :1442-1458
[24]   Comparison of semiparametric and parametric methods for estimating copulas [J].
Kim, Gunky ;
Silvapulle, Mervyn J. ;
Silvapulle, Paramsothy .
COMPUTATIONAL STATISTICS & DATA ANALYSIS, 2007, 51 (06) :2836-2850
[25]   Semiparametric estimation of the error distribution in multivariate regression using copulas [J].
Kim, Gunky ;
Silvapulle, Mervyn J. ;
Silvapulle, Paramsothy .
AUSTRALIAN & NEW ZEALAND JOURNAL OF STATISTICS, 2007, 49 (03) :321-336
[26]   Extreme correlation of international equity markets [J].
Longin, F ;
Solnik, B .
JOURNAL OF FINANCE, 2001, 56 (02) :649-676
[27]  
Nelsen R. B, 2006, INTRO COPULAS, VSecond
[28]   Modelling asymmetric exchange rate dependence [J].
Patton, AJ .
INTERNATIONAL ECONOMIC REVIEW, 2006, 47 (02) :527-556
[29]  
Rodriguez J. C., 2007, J EMPIR FINANC, V14, P401, DOI DOI 10.1016/J.JEMPFIN.2006.07.002
[30]  
Salmon M, 2006, WORKING PAPER SERIES