Asymmetry and Downside Risk in Foreign Exchange Markets
51 Pages Posted: 30 Nov 2000
Date Written: June 27, 2000
A recent paper by Knight, Satchell and Tran (1995) suggested that the double gamma distribution may provide an eective means of modelling asymmetry in financial data. This paper evaluates that claim in the context of the conditional distribution of exchange rate data. To do this, the model proposed by Knight, Satchell and Tran is first extended to incorporate conditional heteroscedasticity and is then applied to ten exchange rate series covering mature and emerging market countries.
A second contribution of this paper is to highlight the link between the double gamma distribution and the measurement of the second lower partial moment (or semi-variance). It is shown that the conditional semi-variance of a series can be easily calculated from the parameters of the double gamma distribution. The resulting empirical performance of the double gamma model is found to be mixed when compared to a symmetric GARCH-t model applied to a range of foreign exchange rates. Estimates of conditional downside risk based on the double gamma model were then constructed for each series. The results for the Malaysian Riggit, Zimbabwe Dollar and the Korean Won demonstrate the extreme downside volatility experienced by these countries during the recent emerging markets currency crisis.
Keywords: Double-gamma, skewness, lower partial moments, garch
JEL Classification: G11, C22
Suggested Citation: Suggested Citation