Inflation Dynamics in Uganda: A Quantile Regression Approach

This paper considers the measurement of inflation persistence in Uganda and how this has changed over time. As the data does not follow a normal distribution, we make use of the quantile regression approach to investigate how various shocks may affect the rate of inflation within different quantiles. The measures of inflation include headline inflation, the central bank's measure of core inflation, and an alternative measure of core inflation. The results suggest that while a unit root is found in many of the upper quantiles of headline inflation, there is evidence of mean reversion within the lower quantiles. In addition, we find higher levels of persistence after 2006 and during the inflation-targeting period. When considering the degree of persistence in the central bank's measure of core inflation, the results suggest that there is a unit root in this measure during the inflation-targeting period. In addition, the alternative measure of core inflation, which is derived from a wavelets transformation, provides similar results. However, this measure is less volatile and more correlated with headline inflation. All the results suggest that large positive deviations from the mean would influence the permanent behaviour of inflation, while small negative deviations are relatively short-lived.