Bamanga, Muhammad A. Musa, Umar Salihu, Audu Udoette, Ubong S. Adejo, Valli T. Edem, Offiong N. Bukar, Hyariju Udechukwu-Peterclaver, Chidinma T.
Year of Publication:
[Journal:] CBN Journal of Applied Statistics [ISSN:] 2476-8472 [Publisher:] The Central Bank of Nigeria [Place:] Abuja [Volume:] 7 [Year:] 2016 [Issue:] 1 [Pages:] 147-169
This paper examines the relationship between inflation and inflation uncertainty in Nigeria. It attempts to test whether the Friedman's hypothesis - that a rise in the average rate of inflation leads to more uncertainty about future rate of inflation - holds for the country. The monthly inflation data spanning the period 1960:1 to 2014:07 was used. Inflation uncertainty was modeled as a time varying process using a GARCH framework. Exponential Generalized Autoregressive Heteroscedasticity (EGARCH) complemented by seasonal ARIMA (2, 0, 2) (0, 0, 1) was employed to model the inflation uncertainty. Given that inflation series display structural breaks, this was tested and found to be significant which was accounted for in the model. The EGARCH fitted our data better than the symmetric GARCH model. The bivariate Granger Causality test was performed on inflation and its uncertainty; it showed that inflation causes inflation uncertainty in Nigeria. The fitted EGARCH model found strong support for the Friedman's hypothesis.