Estimating potential output, core inflation, and the NAIRU as latent variables

This article proposes a new method to obtain estimates of the NAIRU, the core inflation rate, and the trend investment rate for the United States using an unobserved components model that is compatible with the usual decomposition of real gross domestic product into trend and cycle. The model includes Okun's law, a forward-looking Phillips curve, and an accelerator-type investment equation and accounts for some volatility breaks in two components. The unknown parameters in the model are estimated by maximum likelihood using a Kalman filter initialized with a partially diffuse prior, and the unobserved components are estimated using a smoothing algorithm. Our results show that the output gap is positively correlated with the deviations of the investment rate from its trend and the inflation rate from core inflation, and negatively correlated with the deviations of the unemployment rate from the NAIRU.