RT Journal Article
SR Electronic
T1 Predicting Short-Term Eurodollar Futures
JF The Journal of Fixed Income
FD Institutional Investor Journals
SP 47
OP 61
DO 10.3905/JFI.2009.18.4.047
VO 18
IS 4
A1 Choong Tze Chua
A1 Krishna Ramaswamy
A1 Robert A Stine
YR 2009
UL https://pm-research.com/content/18/4/47.abstract
AB We propose and illustrate a structural model for the forward curve produced by Eurodollar futures contracts. Our model provides a three-part functional decomposition of the forward rate: a long-term, unconditional component, a maturity-specific component, and a date-specific component. The maturity-specific component captures preferred investment habitats, and the date-specific component captures shocks to expectations of future spot rates. These functional components (modeled with exponential basis functions) of the decomposition aggregate to an arbitrage-free representation of the underlying stochastic process that drives the evolution of the Eurodollar forward curve. We demonstrate the use of this approach by fitting this model to yields over the period 12/9/1981 to 1/28/2008. The estimation is accomplished by using a Kalman filter to determine the underlying representation. The estimated yield curve provides better out-of-sample predictions than the standard random walk model in forecasts over various horizons. We further show the profitability of a trading scheme that chooses futures positions based upon the anticipated forward curve.TOPICS: Futures and forward contracts, currency, developed markets [Europe], statistical methods