Interest rate futures: Evidence on forecast power, expected premiums, and the unbiased expectations hypothesis

Regression relationships suggested by Fama and French (1987) are estimated to test the forecast power of futures contract prices and evidence of systematic expected premiums. Relative to the contemporaneous spot market price, futures contract prices are found to provide no additional information relevant for forecasting the realized spot price. Consequently, the regression based tests for predictable variation in realized premiums produces no reliable evidence of systematic expected premiums in Treasury bill, Eurodollar, and Treasury bond contract prices. Each of the price series used in the study is found to be nonstationary based on Dickey-Fuller and augmented Dickey-Fuller tests. The maximum likelihood estimation procedure developed by Johansen (1988) is used to test for the existence of a cointegrating relationship between price series for the estimation and parameter restriction tests for the 2 markets for which evidence of a cointegrating relationship is found. Evidence is found of one cointegrating relationship between futures contract prices and the realized spot price for the Treasury bill and Eurodollar markets.