Testing for a Level Effect in Short-Term Interest Rates

There is an extensive theoretical and empirical literature discussing the link between short-term interest rate volatility and interest rate levels. We present an LM based test for the presence of a level effect which is robust to the presence of unidentified nuisance parameter under the null of no level effect. We provide extensive Monte-Carlo evidence on the performance of this test under various DGPs. When applied to data on the 3-month US Treasury Bills rate, the test reports significant evidence of a level effect.