In this study, the forward premium anomaly is revisited. The bias of the forward rate in predicting the future spot rate implies the failure of Uncovered Interest Parity (UIP), the backbone of many economic models in international finance. In analyzing the classical explanations of the puzzle, including the effect of time-varying risk premium and systematic forecast errors, the study takes a unique approach by analyzing the time horizon of the forward rate bias. With daily data of 13-year sample size, ranging from one-month to twelve-month maturity in six major exchange rates, we find a consistent pattern describing that the severity of the anomaly increases with the maturity horizon. The pattern is consistent with time-varying risk premium explanation, suggesting its importance in explaining the puzzle. Further, we test and find evidence supporting the role of risk - measured by the conditional variance of exchange rate return - and its effect over time horizon. When carrying out the test for rolling sub-samples of five-year periods, the majority of the sub-samples conform to our finding of the horizon pattern for the full sample. The exceptions are periods that correspond to long-term swings in the U.S. dollar, from appreciating to depreciating or vice versa, suggesting the effect of peso problem taking over during these periods.
To, Linh, "The Time Horizon of the Forward Premium Puzzle: A New Approach in Examining Classical Explanations" (2007). Economics Honors Projects. Paper 6.
© Copyright is owned by author of this document