A GARCH Model of Inflation and Inflation Uncertainty in Iran

The paper investigates the relationship between inflation and inflation uncertainty using the Iranian data over the period 1959:03 – 2008:02. GARCH models are used to examine this relationship. Granger methods are employed to provide statistical evidence for the relationship between average inflation and inflation uncertainty. Threshold GARCH (TGARCH) models are considered to investigate asymmetry in the conditional variance of inflation. The Component GARCH (CGARCH) models are employed to decompose inflation uncertainty into a short-run and a long-run component by permitting transitory deviations of the conditional volatility around a time-varying trend. This model examines the presence of long memory in the conditional variance of inflation. The findings show that increased inflation raises inflation uncertainty confirming the theoretical predictions made by Friedman. Furthermore, the findings of bi-directional causality support the Cukierman and Meltzer model. Using the standard TGARCH models, the presence of asymmetry is found in the conditional variance of annualized inflation, and finally the evidence of long memory exists in the conditional variance of annualized inflation.