With a little help from my friends

The estimation of dynamic term structure models (DTSMs) turns out to be challenging in the presence of a small sample. It is exacerbated if the sample is characterized by a prolonged period of low interest rates near a time-varying eective lower bound. These challenges all weigh heavily when estimating a DTSM for the euro area OIS yield curve sample. Against this background, we propose a shadow-rate term structure model (SRTSM) that includes a time-varying eective lower bound accounting for the spread between the policy and short-term OIS rate and it also allows for future changes in the eective lower bound. In addition, it incorporates survey information in order to pin down the level of longer-term rate expectations. The model allows to adequately assess short-term monetary policy rate expectations and it generates far-distant rate expectations that are correlated with an estimated equilibrium nominal short rate derived from a macroeconomic model set-up. Our results also highlight the signaling channel of non-standard monetary policy shocks in the run-up to asset purchases based on high frequency identication approach. Our model outperforms DTSM specications without above modeling features from a statistical and economic perspective. We conrm our ndings employing a Monte Carlo simulation.