Detecting time variation in the price puzzle: an improved prior choice for time varying parameter VAR models

This paper compares Bayesian estimators with different prior choices for the time variation of the coefficients of Time Varying Parameter Vector Autoregression models using Monte Carlo simulations. Since the commonly used prior choice only allows for a tiny amount of time variation, less informative priors are proposed. Additional empirical evidence on the time varying response of ination to an interest rate shock is provided for USA. While a major and statistically significant `price puzzle' is detected for the period 1972-1979, the estimated response of ination to an interest rate shock is negative for most other time periods.