An unobserved components model of total factor productivity and the relative price of investment

Abstract This paper applies the common stochastic trends representation approach to time series of total factor productivity (TFP) and relative price investment (RPI) investigate modeling neutral technology (NT) investment-specific (IST) its econometric ramifications on analysis general equilibrium model. The permanent transitory movements in both are estimated efficiently via Markov chain Monte Carlo methods using band matrix algorithms. results indicate that TFP RPI are, each, well represented by a differenced first-order autoregressive process. In addition, their share trend component we interpret as reflecting changes purpose technology. These consistent with studies suggest (1) traditional view assuming NT IST follow independent processes is not supported features (2) improper specification secular may distort estimation inference. Notably, findings provide some guidance minimize effect idiosyncratic misspecifications impulse dynamics propagation mechanisms macroeconomic models.