Abstract:
This paper estimates and compares four versions of the sticky price New Keynesian
model using a Bayesian approach. We estimate an average duration of price contracts
between four and seven quarters in all cases. When we introduce sticky wages we
estimate an average duration of wage contracts to be below three quarters. We find
price indexation to be more important than wage indexation. Finally, the marginal
likelihood criterion ranks both the sticky price and wage model of Erceg, Henderson,
and Levin (2000) and its wage indexation version best.
