A1007
Title: Estimating the natural rate of interest in the US: An accelerating score-driven state space model
Authors: Tibor Pal - University of Salerno (Italy) [presenting]
Giuseppe Storti - University of Salerno (Italy)
Abstract: The aim of the paper is to develop a novel accelerating score-driven state-space model for estimating the natural rate of interest. The proposed model extends the class of score-driven state-space models proposed in another study by assigning a time-varying weight to the conditional likelihood score. The flexible parameters are driven by autocorrelations and cross-correlations of past score innovations, resulting in an accelerated updating mechanism. The model is used to study and estimate the US natural rate of interest in the Laubach-Williams framework, where the IS and Phillips curve relationships become time-varying. In addition to estimating the natural rate of interest, the proposed extensions to the baseline Laubach-Williams framework allow the time-varying nature of the relationships involved to be analyzed and risk appetite to be estimated, thus providing an additional yardstick for monetary policy.
