A Further Examination of the Expectations Hypothesis for the Term Structure

We extend the vector autoregression (VAR) based expectations hypothesis tests of term structure using recent developments in bootstrap literature. Firstly, we use wild bootstrap to allow for conditional heteroskedasticity in the VAR residuals without imposing any parameterization on this heteroskedasticity. Secondly, we endogenize the model selection procedure in the bootstrap replications to reflect true uncertainty. Finally, a stationarity correction is introduced which is designed to prevent finitesample bias adjusted VAR parameters from becoming explosive. When the new methodology is applied to extensive US zero coupon term structure data ranging from 1 month to 10 years, we find less rejections for the theory in a subsample of Jan 1982-Dec 2003 than in Jan 1952-Dec 1978, and when it is rejected it occurs at only the very short and long ends of the maturity spectrum, in contrast to the U shape pattern observed in some of the previous literature.