1 The Effect Of Mis-Estimating Correlation On Value-At- Risk * Vasiliki D. Skintzi 1 , George Skiadopoulos 2 and Apostolos-Paul N. Refenes 3 DRAFT Preliminary and incomplete. Please do not quote without permission from the authors. Comments are very welcomed. 24/10/2002 Abstract This paper examines the systematic relationship between correlation mis-estimation and the corresponding Value-at-Risk (VaR) mis-calculation. To this end, first a semi- parametric approach, and then a parametric approach is developed. Various linear and non- linear portfolios are considered, as well as variance-covariance and Monte-Carlo (MC) simulation methods are employed. We find that the VaR error increases significantly as the correlation error increases, particularly in the case of well-diversified linear portfolios. In the case of option portfolios, this effect is more pronounced for longer-maturity, away from-the-money options. The use of MC simulation to calculate VaR magnifies the correlation bias effect. Our results have important implications for measuring market risk accurately. JEL Classification: G10 Keywords: Value-at-Risk, Correlation, Correlation Mis-Estimation, Monte Carlo Simulation, Variance-Covariance Methods, Model Error. * The authors gratefully acknowledge helpful discussions with Michael Tsiona, and Raphael Markellos. 1 Financial Engineering Research Centre (FRC), Department of Management Science and Technology, Athens University of Economics and Business (AUEB), 47A Evelpidon & 33 Lefkados 113 62 Athens, Greece. vikiski@aueb.gr 2 Department of Banking and Financial Management, University of Piraeus, Piraeus, Greece, Financial Options Research Centre, Warwick Business School, University of Warwick, and FRC, AUEB, Athens, Greece. gskiadopoulos@hotmail.com 3 FRC, Department of Management Science and Technology, AUEB, 47A Evelpidon & 33 Lefkados 113 62 Athens, Greece. prefenes@aueb.gr