IJARP Impact Factor(2018): 4.908

International Journal of Advanced Research and Publications!

MODELLING AND FORECASTING VOLATILITY OF PRICE INFLATION IN ETHIOPIA USING GARCH FAMILY MODELS

Volume 4 - Issue 1, January 2020 Edition
[Download Full Paper]

Author(s)
DEREJE DEMEKE
Keywords

Abstract
MODELLING AND FORECASTING VOLATILITY OF PRICE INFLATION IN ETHIOPIA USING GARCH FAMILY MODELS ABSTRACT Inflation and its volatility is one of the series macro-economic problems in every countries economy. Inflation in Ethiopia is not immune from this problem nowadays and it is vital to model and forecast it. Therefore, this study aimed at modelling and forecasting price inflation volatility in Ethiopia using GARCH family models for the general inflation data, which spans from 1995 to2016. The Lagrangian Multiplier (ARCH -LM) test statistic is used for testing the existence of ARCH effect in the residuals of conditional mean or ARMA (1,1) model and it confirmed the existence of ARCH effect in the price inflation series in Ethiopia. This indicates price inflation in Ethiopia is suffered from volatility problems and applying GARCH family models is relevant and necessary. To model and forecast the price inflation volatility, ARMA (1,1)-GARCH (1,1) was selected as an appropriate model among EGARCH (1,1) and GARCH (1,1) models with GED, normal and t-distributional assumption for residuals. To select an appropriate model, forecasting error measure statistics such as: MAE (Mean Absolute Error), RMSE (Root Mean Square Error) and Uthail’s inequality coefficient were used in addition to well-known information criterias such as: AIC (Akeike Information Criteria) and BIC (Byesian Information Criteria). Moreover, macro-economic variables such as: Broad Money Supply, Exchange Rate and Lending Interest Rate have direct contribution for the price inflation volatility in Ethiopia except Deposit Interest Rate and GDP(Gross Domestic Product). The finding of this study also clearly showed that last shock and volatility had significant contribution to price inflation volatility. Finally, the price inflation volatility was forecasted using ARMA (1,1)-GARCH (1,1) model with GED distributional assumption. The forecast showed the existence of fluctuation of variance which is declining at the end of the study period. This study suggested that, to come up with stable price inflation volatility in Ethiopia, the government as well as concerning bodies must pay great effort to control macro-economic factors of inflation volatility. Key words: ARMA, EGARCH, GARCH, Forecasting, Inflation
References