Investigating the Relationship between Inflation Rates and Inflation Uncertainty in Iran by Using Markov-Switching Regression

This paper studies the relation between inflation rates and its uncertainty by using Markov-Switching regression model and monthly data of consumer price index during 1990:01-2012:09 in Iran. Inflation uncertainty is estimated by using Generalized Autoregressive Conditional Heteroskedastisity (GARCH) model. The empirical results of Markov Switching Auto Regressive (MSAR) model represent the presence of two clearly differentiated regimes over the entire Sample. The first regime corresponds to a high level inflation rate, low volatility, and the second regime corresponds to low level inflation, high volatility. The use of Markov-Switching Regression Model indicates that the increase of inflation rate will be associated to higher uncertainty according to both regimes.