Heavy-Tailed Behavior of High-Frequency Foreign Exchange Changes

• Activity Number: 517
• Type: Contributed
• Date/Time: Wednesday, August 4, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Business and Economic Statistics Section
• Author(s): Ece Oral*+
• Companies: Central Bank of the Republic of Turkey
• Keywords: Stable distribution ; Distribution functions ; Parameter estimation ; high-frequency data ; Foreign exchange rates

Abstract

Numerous techniques in finance rely deeply on the assumption that the random variables under investigation follow a Gaussian distribution. Yet, finance data often depart from the Gaussian model, in that their marginal distributions are fat-tailed. In the literature on the empirical distribution of foreign exchange rates there is now consensus that exchange rate yields are heavy-tailed. In the presence of heavy-tails it is natural to assume that the exchange rate returns are approximately governed by a stable distribution which has been used successfully for modeling stock returns and foreign exchange rates. This paper investigates the distribution of high intra-daily TRY/USD foreign exchange changes. Stable laws are fitted to empirical distributions. Empirical evidence supports that the distributions have tails much heavier than Gaussian distribution.