 |
Financial market participants and policymakers closely monitor the evolution of interest
rate expectations. At any given time, the term structure of interest rates contains information
regarding these expectations. No-arbitrage dynamic term structure models have
regularly been used to estimate interest rate expectations and term premia, but daily frequency
estimates of these models fail to accurately capture the evolution of interest rate
expectations implied by surveys and nancial market instruments. I propose the augmentation
of no-arbitrage Gaussian affne dynamic term structure models (GADTSMs) with
overnight indexed swap (OIS) rates in order to better estimate the evolution of interest rate
expectations and term premia across the whole term structure. I augment the model with
3 to 24-month OIS rates, which provide accurate information about interest rate expectations.
The OIS-augmented model that I propose, estimated between January 2002 and
December 2016 for the US, generates estimates of the expected path of short-term interest
rates, up to the 10-year horizon, that closely correspond to those implied by federal funds
futures rates and survey expectations at a range of horizons, and accurately depict their
daily frequency evolution. Against these metrics, the interest rate expectation estimates
from OIS-augmented models are superior to estimates from existing GADTSMs.
